Qualitative Methods in Social Research Qualitative Methods in Social Research Kristin G. Esterberg University of Massachusetts-Lowell Mc Graw Hill Boston Burr Ridge, IL Dubuque, IA Madison, Wl New York San Francisco St. Louis Bangkok Bogota Caracas Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal New Delhi Santiago Seoul Singapore Sydney Taipei Toronto (lontents PiiTace xiii ( IIAPTER 1 What Is Social Research? Some Practical and Theoretical Concerns What Is Social Research? 1 Why Do Research? 3 I Vveloping a Sociological Imagination 4 I lie Research Process: Moving Between Theory and Data 5 I \ isitivism: Traditional Approaches to Social Research 10 » challenges to Traditional Ways of Doing Social Science 11 "You Are Here": Locating the Self in Social Research 12 Naturalism 13 Social Constructionist and Interpretive Approaches 15 I vminist and Other Critical Approaches 17 Critical Social Research 17 Feminist Research 18 Postmodernism 20 Making Claims About Paradigms 22 Questions for Thought 22 llxercises 22 Suggestions for Further Reading 23 CHAPTER 2 Strategies for Beginning Research 25 Getting Started: Where to Begin? 25 Generating Ideas 26 Deciding What to Research 2H How Do You Turn a Topic into a Question? 2!) Is the Topic Interesting? 30 Whose Perspective Should You Take? 30 Is the Research Feasible? 31 Is the Research Worth Doing? 32 Developing a Research Strategy 33 Different Traditions, Different Starting Points 33 Types of Research Strategies 35 Reading the Literature 37 Evaluating Web Sites 38 The Research Process in Review 39 Question for Thought 40 Exercises 41 Suggestions for Further Reading 41 CHAPTER 3 Ethical Issues 43 Sleeping Well at Night: A Practical Ethics 44 Ethical Codes 44 Confidentiality and Informed Consent 45 Controversies in Social Research 46 Milgram's "Obedience" Study 46 Humphreys' "Tearoom " Study 4 7 Feminist Approaches to Research Ethics 48 Institutional Review Boards 49 Ethical Concerns in Qualitative Research 50 Some Problems of Confidentiality and Intimacy 50 Some Problems of Informed Consent 51 Overt Versus Covert Observation 52 Practical Concerns in Protecting Privacy 53 Talking About Ethical Issues 55 Questions for Thought 5 5 Exercises 56 Suggestions for Further Reading 56 CHAPTER 4 Observation: Participant and Otherwise 57 Ethnography 58 Participating and Observing 60 The Self as Instrument 61 I I.ml me Aliniil Sellings :uul Srln Iiiij', ;i Site 62 I / his iiul Gaining Access 68 / 'crclopiug Relationships 69 • nhii'iiting Informants 70 / developing a Cover Story 70 /!<•, oining Invisible 71 I hulling with Emotions 72 1 - ■ it in)' Meld Notes 72 / Vf 'doping Observational Skills 72 When to Write? 73 What to Write? 74 I low to Write? 77 i ■ .ivmg the Field 78 i •" using Your Observations 79 A /< (king Meaning 7 9 < 'oming to See Multiple Realities 80 1 >u.-siuiiis forThought 80 I >, I, ISC.V 81 uiry\iions for Further Reading 82 < IIAPTER 5 Interviews 83 Interviewing as a Relationship 84 I v | h-s of Interviews 85 Structured Interviews 85 Semistructured Interviews 87 Unstructured Interviews 89 < loseness and Distance in Interview Situations 90 I 'i (paring for the Interview 92 Deciding Whom to Interview 93 Preparing an Interview Guide 94 Deciding What Kinds of Questions to Ask 95 Structuring and Ordering Questions 96 Making Questions Open-Ended 98 Pretesting Your Interview Guide 100 Before the Interview 100 Establishing a Location 1 01 Constructing a Face Sheet 101 Deciding What to Bring 101 Dressing Appropriately 102 During the Interview 102 Warming Up 103 Keeping the Conversation Rolling 103 Thinking on Your Feet: Follow- Ups and Probes 104 Speaking and Keeping Quiet 104 Communicating Nonverbally 105 Keeping on Track 105 Taping and Taking Notes 106 After the Interview 107 Writing Field Notes 107 Transcribing and Making Sense 107 Group Interviews 108 The Role of the Focus Group 108 The Role of Moderator 110 Ethical Issues in Conducting Focus Groups 111 Recording the Focus Group 111 Making Meaning 112 Questions for Thought 112 Exercises 112 Suggestions for Further Reading 113 CHAPTER 6 Unobtrusive Measures: Analyzing Texts and Material Artifacts Physical Traces 116 Material Artifacts 117 Documents and Records 121 Public Records 121 Documents and Private Papers 122 Primary and Secondary Sources 123 Media Accounts 124 Electronic Texts 125 Historical Research 128 Doing Historical-Comparative Research 129 Gaining Access to Historical Materials 130 115 ' I \hiittni)[ Sunn es I 3 I ■ I. .iiiiif Meaning IM /,'/ Ill.Hlf'Jll 133 • ■ . . Ill ■ i.'ic. /.>/ I'inlliei>■ Udlding 134 i",c <>l Action Research 135 .11- in dI A( tion Research 137 I'.niii i/hiioiy Action Research (PAR) 137 I, iiinnsi Action Research 139 ' r.tliitiiion Research 140 i> ■ ii. Ii Ioi Whom? 140 I * * 11" > \vc ■ i merit 141 I I,. i'i,ess of Action Research: Look, Think, Act 142 / «'.i/.'- Identifying Stakeholders and Formulating ,i ( ollective Problem 142 ' limk: Exploring the Problem 145 \< I I )cfining an Agenda for Action 145 I' ii nie ilie Results 146 .....■■ I'i oblems and Opportunities in Action Research 146 '•litiivig Power 147 I i-iidiiig Groups 148 . Ii ting Ethically 148 1 '<■. Hum lor Thought 149 ' • 149 ." r, -.ihnisjor Further Reading 149 « 11 \l" I !• R 8 Making Sense of Data 151 h I il in;; Meaning Out of Qualitative Data 152 M.in.ij'jng Data 153 Srfurating Different Types of Data 153 keeping Data in Chronological Order 154 ' organizing by Topic or Document Type 154 Making a List or Logbook 155 ( hoosing Between Computer and Hard Copy 156 Keeping Multiple Copies 156 « "King Intimate with Your Data 157 Coding 157 Open Coding 158 Development of Themes 159 Focused Coding 161 Maps and Diagrams 162 Memos 164 Procedural Memos 165 Analytic Memos 165 Asking Questions 166 Developing an Analysis 168 Looking for Patterns 168 Comparing Cases 168 Building Typologies 169 Getting Stuck and Unstuck 170 Content Analysis 171 Semiotic Analysis 172 Grounding the Analysis 173 Shortcomings in the Research 174 Negative Cases 174 The Null Hypothesis Trick 175 Triangulation 176 Computer-Assisted Data Analysis 176 Question for Thought 179 Exercises 179 Suggestions for Further Reading 180 CHAPTER 9 Narrative Analysis 181 What Is a Narrative? 182 The Structure of Stories 183 Other Structures of Stories 185 Structure, Story, and Social Context 187 A Focus on Structure 188 The Social Contexts of Stories 188 The Story Itself 190 Telling the Story in Other Ways 192 Some Practical Advice 192 Evaluating Narratives 195 Questions for Thought 196 Exercises 196 Suggestions for Further Reading 197 I II Mil It 10 WriliiiK Alioiil Itfscanh I *)*) ' • Mim' r, .1 I'lot CSS 200 '■•mil,; I hwu to Write: lust Do It 201 H. nimm): Writing 202 '■■um/: Notes 202 I in, Im/; I >ireition: Making a Writing Plan 203 II nfiin: a First Draft 203 rciiij; mill Fditing 203 I'nu'lieiiiling 204 ■ •..Ii. ii,.-, .nid Venues 205 1 1 ImI W.ivn to Structure Qualitative Research 207 / iile and Abstract 207 liitnuliiction 208 ' iteitilnre Review 210 in 11 rs of Data and Methodology 211 I he Main Body: Findings 213 '.iimiiiiiry and Conclusion 214 I ist of References and Appendixes 214 I I il inj'. Writing Vivid 215 1 mIi. i ( , i mi res of Qualitative Research 216 ■ ii.nis In/Thought 217 ..... •.n.iif.s jar Further Reading 218 S|'|M'iuli.v A: American Sociological Association Code of Ethics S|p|' 11 ■ m -i 11 in our family life. My life and work are much richer because of it. finally, thanks to my students, at both the University of Missouri-I .msas City and the University of Massachusetts at Lowell, My Kansas ' 11 v students generously served as "guinea pigs," allowing me to initially . Ii'velop my ideas about teaching qualitative methodologies. My students .ii I owell continually push me to be a better teacher. I thank them all. ACKNOWLEDGMENTS A number of people have helped make this book a much different—and better—book than I could have managed on my own. Thanks, first, to Serina Beauparlant, an extraordinary editor. Serina is a grueling taskmaster and supportive friend. I'm thankful to have worked with her on this book. Kazuyo Masuda and Sue Bergmeier generously read portions of 1 What Is Social Research? Some Practical and Theoretical Concerns WHAT IS SOCIAL RESEARCH? What is social research? How does it differ from, say, journalism, or philosophy, or fiction, or any other way of knowing about the world? How does i|ualitative research, which this textbook focuses on, differ from quantitative research? What do social researchers do? Sociologists answer these questions in many ways. These answers often reflect deep philosophical differences about the nature of social reality and the ways in which one should study it. Sociologists who prefer quantitative methodologies tend to argue that, unless researchers use something called "the" scientific method and follow the same kinds of rules that natural scientists (such as chemists) use, it isn't really social science research. Other sociologists believe that social science research is fundamentally different from the natural sciences. They argue that social research is primarily a matter of interpretation. In their eyes, the most important goal of social research is to investigate and illuminate how humans construct social reality. I argue in this book that there are many different ways to do social research, with many different aims. What all these methods share is the goal of learning something about the social world, however that world is construed. While social scientists may disagree, sometimes heatedly, about the nature of social reality and the best ways to study it, they all agree on the importance of understanding the social world. Although I was trained in a quantitative tradition, modeled after the natural sciences, I have come to 1 adopt an interpretive approach in my own research and wilting Y. ( I nit skill to learn. ('onsider some of the difficulties. Let's imagine that you have been ob-* i ving children on a playground for several months. You have visited sev- ■ id times a week for an hour at a time, and you have tried to take accurate mil detailed notes about what you have seen. You have observed children .Lung many things: playing hopscotch and soccer and four-square, chasing ..ne another, talking, arguing, yelling, crying. You have many pages of notes ili.il document, in detail, a slice of children's playground life. How do you i h.n make sense of it all? How do you begin to identify larger social pat-11 'i ns? How do you move from your notes and observations to a sociological niiiilysis? That's what this book aims to help you with. WHY DO RESEARCH? I'.ople conduct research for many reasons. Some do it because it's fun— iliry enjoy the challenge of gathering data and trying to make sense of it. I >uing research is a process of exploration, a way of finding out things that ihey're interested in. Other people conduct research because they have to, perhaps as part of their degree requirement or their work. Many people have jobs that require some type of research skills. Social workers, for example, may need to do social research to find out if a particular program or policy is effective. They might want to know if welfare-to-work initiatives that aim to move poor women into the labor force actually work. Do they actually help poor women move out of poverty? Others, like community organizers, social policy makers, or teachers, may need to read reports compiled by prolessional researchers. Teachers m;iy need lo know, lm enample, if whole-language reading strategies work better than plinnii s lm -.unie children. Even if they do not want to do the research themselves, they need to know what others have found. They also need to know how to evaluate research, rather than simply accept it at face value. If several studies suggest that whole-language reading strategies work better and several others recommend phonics-based approaches, the teachers need to know how to reconcile what seem to be conflicting results. Some researchers are motivated by a sense of social justice. They want to right what they see as social wrongs, and they want to use social research to aid in that effort. For example, Ronnie Steinberg has conducted research on pay inequities in order to help close the gap between men's and women's pay. She describes herself as a feminist social scientist who does advocacy work on behalf of women (1996, p. 225). Others are motivated by a deep curiosity about the social world. Although basic research, which is aimed at creating knowledge for its own sake, may not have an immediate, practical purpose, it helps us to understand social life. For example, sociologist Jack Katz (1996) analyzed how families interacted in a Paris fun house in order to understand the social construction of humor. Although this research may not have an immediate application, knowing more about how people construct humor may—or may not—ultimately have some practical use. Like Katz, many social scientists conduct research because it's a way of learning about things that interest them. DEVELOPING A SOCIOLOGICAL IMAGINATION We shouldn't think of social research methods as merely a set of cookbook procedures for obtaining information. Whether qualitative or quantitative, social research methods are intertwined with theoretical concerns. When you try to understand the social world, you are developing what C. Wright Mills (1959) called a sociological imagination: the ability to see individual issues within a larger social context. Developing this sociological imagination involves theorizing. Sometimes, students groan or their eyes glaze over when I mention theory. People often think that theory is necessarily boring or arcane and clearly not useful in the "real" social world. But the ability to theorize is a highly useful skill. Life would be very confusing without the ability to theorize. In fact, you theorize all the time—you just don't think about what you're doing in that way. Any time you try to understand the world around you, you are theorizing. You have theories for why your professors act the way they do and what • ill happen if you turn in ;m assignment late. You have theories tor why ..mi people get paid more than others and why some people go on to col-I i" ,ni,e used by these theorists laborious, perhaps difficult to understand, il..-.. are examples of theories that provide grand, overarching explanations ..I ..ii ial phenomena. Although these types of explanation are certainly i li. . .1 v, they are not the only kind. Another way to think about theory is as a i. n v about some event or some piece of the social world. A theory helps pi.ivide an explanation for a whole class of events. Some theories are highly il.-.iiacl and difficult to understand; others are not. If you think about theo- ii. •. in this way, you can see that you use them all the time. I or example, let's say your parents immigrated to the United States b. l-.ie you were born. They have ways of doing things that they brought Ii.hii their home country and ways of doing things they learned here. Be- ■ .lice you were born in the United States, however, you feel clearly Ameri- ■ in Yet you also have strong ties to your ethnic community. Sometimes, v mi parents seem too strict; other times, they seem just right. You probably have developed theories about why they act the way they do. One way of ■ \ plaining your parents would be to theorize about why, individually, they i. I the way they do. You might see your mother as very strict in comparison wiili some of your friends' mothers, and maybe that is simply part of her p.-isonality. Or you might think about whether your family shares some .. mimonalities with other immigrant families. In that case, you might want i. 11 heorize about how the experience of immigration affects family life. This would be an attempt to explain a whole class of families. I III RESEARCH PROCESS: MOVING BETWEEN THEORY AND DATA He I ore you can begin to conduct social research, you need to consider the itlationship between theories and the empirical world. The empirical world is the world of the senses: the world you can see, hear, smell, touch, and (less frequently considered in the social sciences) taste. Traditional social research draws on the model of a natural scientist conducting research in a labora-lory. In this tradition, often called the "scientific method," the main goal of Stage 1 _ Develop theory a Stage 2 Operationalize /\ Develop testable Choose sample and hypotheses _____ research design Stage 3 Collect and analyze data Stage 4 Disseminate results FIGURE 1.1 A Deductive Approach to Research social research is theory construction and, most importantly, theory testing. Conventional social research uses deductive reasoning. That is, you begin with a theory and then deduce logical extensions of it, called hypotheses, that you can test. The process of deductive reasoning is usually described as having several stages. The first stage involves developing a theory, usually based on the body of research that other scholars have already conducted. The second stage involves operationalizing the theory—that is, putting it in a testable form—by developing hypotheses and choosing a representative sample and a research design. The third stage involves actually carrying out the research: collecting data and conducting analyses. If the results of the test confirm the hypotheses, then the theory is considered more plausible. If not, the theory needs to be reconsidered and further research conducted. The final stage involves writing the results up and disseminating them either in a journal or book or in an oral presentation at a professional conference. Figure 1.1 summarizes the deductive reasoning process. For example, Phyllis Moen wanted to investigate what factors might affect mothers' well-being. One theory suggested that when mothers work outside the home they experience role strain from being pulled in too many conflicting directions, and thus report greater stress (stage 1). Moen decided to test this hypothesis by measuring psychological distress among a sample of Swedish parents (stage 2]. She found that the mothers reported much An.ily/e evidence Develop theory---- ■ K.I. IKK 1.2 An Inductive Approach to Research I. v. distress in 1981 than they had in earlier periods, even though they were wioic likely to work outside the home in the later time period. Thus, she ■ 'imluded that role strain isn't useful in understanding the effect of 'mien's paid employment on their well-being (stage 3). She published this i. ..mix h in her 1989 book Working Parents (stage 4). In qualitative research, investigators typically are less concerned with i In. kind of theory testing. Qualitative research often uses inductive reason-in,- That is, rather than beginning with a particular theory and then looking ii i lie empirical world to see if the theory is supported by "facts," you begin I >v examining the social world and, in that process, develop a theory consis-i. hi with what you are seeing. This approach, illustrated in Figure 1.2, is ■ 'lien called a "grounded approach" (Strauss and Corbin 1990). For example, Susan Walzer (1998] also was interested in studying moth-' i'.' well-being, but in a very different way. She wanted to understand the I'uieess by which women and men become mothers and fathers—that is, In i\v couples negotiate transitions to parenthood. Instead of beginning with i ilieory to test, however, she selected a sample of 50 new parents to inter-Mew in depth. As she analyzed the interviews, she realized that the experi- ■ in es of the women seemed similar, as did the men's experiences. Based mi lier observations, she began to think about the different cultural mean-ni)',s of parenthood for men and for women. The theories she developed to ii v to explain this gender differentiation arose from the empirical evidence .lie gathered. There is a long history ol grounded research wllliln mk inlogy. In the 1920s and tor decades after, the University of Chicago scivcil us n center of qualitative field research. Sociologists there saw the urban selling as a "social laboratory" for social scientists and social reformers (Park 1%7). The Chicago School, as it was called, trained students to go out into the world and study the people and settings they encountered. Researchers trained in the Chicago School investigated, among other topics, Italian Americans living in an urban slum (Gans 1962), medical students (Becker, Geer, Hughes, and Strauss 1961), and marijuana smokers (Becker 1963). These scholars were encouraged to see how their empirical research could be "integrated with" social theory (Shaffir and Stebbins 1991, p. 9). At the same time, prominent anthropologists like Margaret Mead, Franz Boas, and Bronislaw Malinowski were developing procedures for fieldwork within the field of anthropology. This kind of research—intensively studying a specific social group by observing the group in its natural setting—is known as ethnography and sometimes as participant observation. Researchers who do this work are called ethnographers. Women scholars were active in the early years of Chicago School sociology—and even earlier. Harriet Martineau, for example, was one of the founders of sociology (one whose role has been little discussed until recently). Her book Society in America, published in 1837, is considered by some to be one of the earliest examples of ethnographic research (Deegan 1991; Reinharz 1992). Later, in the period 1890-1920, a number of women scholars such as Jane Addams, Edith Abbott, and Sophonisba Breckenridge actively engaged in qualitative research. These women, many of whom were involved in applying the tools of sociology to the pressing problems of the day, had an impact on the development of sociology within the United States. In fact, Mary Jo Deegan, who has studied extensively the role of women in the history of sociology, argues that women scholars in the settlement movement both predated and actively shaped the contours of the more famous male scholars of the Chicago School (Deegan 1991).* Sometimes, researchers move back and forth between inductive and deductive reasoning. Another way to think about the research process is as an ongoing dialogue between theoretical concerns and empirical evi- *The settlement houses, established during the late nineteenth and early twentieth centuries, were centers of moral reform and progressive activism. Women involved with the settlement house movement were involved in a number of activities aimed at social improvement, including work on labor legislation (such as that establishing the 8-hour workday for women and children), housing, and public health and sanitation. The most famous was Hull-House in Chicago, established by Jane Addams in 1889. At Hull-House, women activists and scholars lived and supported each other. Some of the most important early women sociologists were associated with Hull-House and the settlement movement (Deegan 1991). .1. in >' II theories arc stories about the way the world (or some portion ol it) ». "i I'., (hen they are always in a stall' ol revision, and there are always other, ili. i ii,ii ive stories that could be told. In this sense, we are never really done ill. .ui/nif;, anil we can rarely reject theories out of hand. Instead, we need i.. 111111 k about the multiple stories that might be told.Atthe same time, the i. 11 ii r,!, ol telling stories alerts us to different features of the social world. Mn hael Burawoy describes a process he calls "theory reconstruction" I Hin .iwuy et al. 1991). He argues that one of the goals of research is to ex-i. ii.I aiid improve existing theories based on an awareness of features of the • mimical world that aren't explainable by current theories. When we do i ■ ,. .in h we find things that, based on our theories, we didn't expect to find. Mm instead of interpreting these puzzles as a failure of the theories (and a in . 11 tu reject them), he argues, we should use these "failures" to improve ..in theories. Whatever you think about the relationship between theory and data, ■ i.il research still entails some kind of movement between theoretical and • mimical concerns. Different ways of thinking about the research process involve different paradigms, or worldviews. Rather than have your world--1. w remain implicit, or understood, it is much better to make it explicit. I In i hoices researchers make about paradigms shape the research strategies i In-y t hink they should use. These are partisan choices, and they reflect the 11.nning, sensibilities, and beliefs of researchers. That's why, as researchers, \.....iced to think reflexively—that is, to think about who you are and what \. mi beliefs about the social world are—in order to make these decisions. As basic worldviews, paradigms represent beliefs about the nature of i. alily and the ways in which we create knowledge. Scientific paradigms iiieinpt to answer a number of questions about social research: What is the purpose of social research? Should social research aim to improve the social world or merely comment on it? What is the nature of social reality? ■ Is there an objectively "knowable" world out there, or is all knowledge subjective? What constitutes a good explanation or theory? "■' How does one evaluate any particular theory? Paradigms are not provable. That is, you cannot prove that one paradigm is essentially better than another. They are, essentially, matters of faith. But paradigms shape the methodological choices you make and the irlalion-ships you see between theory and data. In this chapter, I'll discuss five different research traditions. The first one, called positivism, has been the dominant tradition in sociology since World War II, especially in quantitative research. A number of new traditions, in addition to the earlier tradition of field research developed by the Chicago School, have developed in opposition to this way of doing research. I'll discuss four alternatives: naturalism, social constructionism, feminism and critical approaches, and postmodernism. What's most important is not that you remember the labels, but that you see how they point you toward different directions in your research. POSITIVISM: TRADITIONAL APPROACHES TO SOCIAL RESEARCH Traditional approaches to social research are based on a paradigm known as positivism. In this tradition, the goal of social research is to discover a set of causal laws that can be used to predict general patterns of human behavior. Prediction is closely related to social control. If you can predict people's behavior, then you can also find ways to control it. Early positivists like Auguste Comte believed that sociology could become a "positive" science of society. By discovering the laws that governed social behavior, sociologists could develop policies that would improve, or even perfect, society. The paradigm of positivism assumes that the social world is inherently knowable and that we can all agree on the nature of social reality. The social world thus has a regular order that social scientists can discover. Knowledge is created by deductive logic: finding ways to operationalize and then test social theories. Explanations in the form of causal reasoning are taken as "true" when they have no logical contradictions and are consistent with observed facts (empirical evidence}. In this tradition, there is a sharp break between scientific ways of knowing and other ways of knowing (such as religion, intuition, or magic). For example, suppose you have a theory that groups of oppressed people will protest when social conditions are at their worst. You go out and measure the social conditions (such as unfair laws, a lack of jobs and housing, and high arrest rates) of different groups of people. And you have good measures of protests—riots, demonstrations, and the like. What you find is that people seem to protest most not when social conditions are at rock bottom, as your theory predicted, but when things are improving. In this case, the empirical evidence (the measures of protests and social conditions) log- U ill iilly coiilraJicls your llirory. In I In- posit ivist tradition, you would have to i.'in t your theory and come up with a new one. (This is, in tact, what many ,nt i;il researchers did.) Social scientists who work within this tradition argue that social re-■.e.iivli must be value-free and objective. Social researchers must somehow lire themselves from the social and cultural values that govern other kinds . .1' human activity. They must transcend personal biases, prejudices, and values and remain neutral toward their object of study. CHALLENGES TO TRADITIONAL WAYS 01 DOING SOCIAL SCIENCE There have been many challenges to the positivist tradition, on many grounds. For one thing, studying humans is different from studying other aspects of the natural world because human behavior isn't mechanistic. Humans have the capacity to reflect on their actions. In fact, when you study people, chances are they're going to change their behavior—even subtly— I ust because you are focusing on them. Molecules or atoms and other aspects of the physical world don't, by and large, do this. And unlike many features of the physical world, human behavior is very context sensitive. Tims, if you bring people into an experimental laboratory to study them, their behavior will be different from what it would be if you observe them in their homes or workplaces or other natural settings. In addition, human reality is multifaceted. Humans can express themselves through art and literature and other forms of self-expression in addition to more goal-directed forms of behavior. Thus, some social scientists argue, it doesn't make sense to study humans using the same methodologies that physicists or chemists might. Furthermore, in social research, humans are the researchers as well as the objects of study, which means that pure objectivity is impossible. We have a vested interest in what we study. As Dorothy Smith argues, "In the social sciences the pursuit of objectivity makes it possible for people to be paid to pursue a knowledge to which they are otherwise indifferent" (1987, p. 88). We are not indifferent to what we study! In fact, if we look at the ways in which social researchers have developed their theories and framed their research projects, we can see that these reflect the interests and priorities of the researchers. And because most researchers and theorists have come from the upper social classes, it's no surprise that much social research reflects the views of those people who have more power in society. Theorists like Nancy I lartsock (1987), Sandra I hm«1111>■ ( I'M,), and Patricia Hill Collins [1990) have argued that, if reseati In is I1 their investigation from the perspective of dispossessed groups, they will end up with a very different perspective than if they begin from the perspective of the wealthy and powerful. So, for example, if you begin from the standpoint of poor women, you probably wouldn't develop theories of welfare rooted in the assumption that poor women are "lazy" or don't want to work. Rather, you probably would begin with an entirely different understanding of the problem of welfare. For example, you might begin by looking at the realities of poor women's lives, such as the low wages available or the difficulty in gaining decent child care or health care. Challengers to traditional ways of doing social science argue that all knowledge is created within human interaction. Who we are shapes the kinds of theories we create and the kinds of explanations we offer. Instead of assuming that objectivity is possible, then, we need to be reflexive: We need to develop an understanding of how our positions shape the research topics we choose and the methods we use to study the social world. Literally, what we see is shaped by who we are. Laura Ellingson (1998] discusses these issues in her research on cancer survivors. A survivor of bone cancer herself, she argues that her experiences gave her crucial insights into the worlds of cancer survivors and clinic staff. Not only did she have a more thorough "technical understanding" of the clinical setting than other researchers who did not share her background, she also had an increased empathy for the patients she was studying. "Although no one can fully understand another's experience," she argues, "I come closer to putting myself in the place of another than one who has never known life threatening illness" (1998, p. 497). Because of her experiences, the research process itself was not without pain. For example, when she observed a patient ask her clinician about an endoscopy, she experienced nausea and gagged, remembering her own endoscopy years earlier. The way in which Ellingson wrote up her research reflects her position. She interspersed her report with memories of her own cancer treatment. "YOU ARE HERE": LOCATING THE SELF IN SOCIAL RESEARCH "You are here." Have you ever seen a map at, say, a mall or a tourist attraction, that labels where you are in relation to other places of interest? Study- ni)', these maps, you can limine out where you arc and how to gel to where ...u want to go. You need to do something similar in social research. Before ..ii liegin your research project, you have to consider where you stand on a .......I mi of important issues: What are your own biases and preconceptions? What are your own investments in particular issues and in particular ways of seeing the world? What do you already think you know, and how do you know it? in lead of thinking of yourself as a neutral, disinterested observer, think .1" ml the connections that you bring to what you plan to study. We'll con-ii lei lour research traditions that encourage you to ask these kinds of ques- ......v naturalism, social constructionism, feminist and critical approaches, .inI postmodernism. M VI URALISM < 'ne ,,1 the most influential approaches in qualitative research, especially as ■ -Ii.lik ted by many anthropologists and other field researchers, has been • i.iinnilism. The goal of naturalistic forms of inquiry is to present the lives 11111 perspectives of those being studied as faithfully as possible. Naturalistic i. search is often conducted in a particular geographic place (Gubrium and I lolstein 1997). One of the best-known examples of this is Tally's Corner, a iii.lv of urban Black men whose lives centered around a particular street .....ur (Liebow 1967). Elliot Liebow wanted to understand the social lives . .1 these men on their own terms. So, rather than begin with a concept like . Ii Imquency," Liebow began by hanging out with the men to understand, as In ■,( he could, their experience. The classic naturalist image is that of the field researcher who goes out into distant social worlds (either literally or figuratively) to study the people u iihin them. In naturalistic inquiry, the researcher attempts to observe as ■ ai efully and accurately as possible and to present the stories of those being ■.ii ii lied in their own voices. As Norman Denzin notes, this research is r.iounded in the "behaviors, languages, definitions, attitudes, and feelings of those studied" (1989, p. 71). To accomplish this, the researcher has to ilevelop close, personal, and empathic relationships with those being stud-i" I, she or he has to become fully engaged with their world. mill positivisl "' out there Although there arc a lew similarities between naluuillsUi research (the assumption, for example, that there is a "real wmli that the researcher can document), there are many more differences. In naturalistic inquiry, the goal is not to abstract a few concepts and to determine the causal relationships among them, but to understand the social world of those being studied. The social context is crucial in naturalistic research. Typically, naturalistic researchers immerse themselves in their field settings, often living among those being studied for long periods of time. Rather than relying on impersonal methods (such as surveys or questionnaires), naturalistic observers rely on their own powers of observation or on in-depth personal interviews to collect data. Although naturalism remains an important method within qualitative research, it is being supplanted by other paradigms. Critics have identified several problems with naturalism (Denzin and Lincoln 1998). For one thing, this tradition assumes that researchers can accurately portray the concerns and issues of those being studied. But the way in which they produce their social research privileges the researchers' accounts. Once they have collected their data, they still must present the final story. Because researchers have the final say, as it were, their account carries more weight than the accounts of those being researched. Researchers get to choose what to present and what to leave out and how to portray those being studied. The people who have been made the objects of study might disagree with the conclusions, but because they are not writing the final report, their disagreements may not be aired. Critics also argue that it is impossible for naturalistic researchers to produce objective accounts. Rather, naturalistic reports are always filtered through researchers' perspectives. Also, when observing, researchers cannot attend to everything at once. For example, try to observe and write down everything that is going on in a classroom for just 10 or 15 minutes. You probably will not be able to see or hear everything or to understand all of what you see, and you certainly cannot write down everything. Therefore, you have to pick and choose among what you think is important. So it is with naturalistic research. The naturalistic observer selects what she or he thinks is important and, in this way, creates his or her own version of reality. More fundamentally, the naturalistic perspective assumes that there's a social world "out there" that can be faithfully studied and reproduced. Other critics, such as those we will consider next, argue that all social life is constructed. Everyday life, in this view, is created through social interaction, and the activity of "conducting research" is no different. Thus, we need to pay attention to the researcher and to the social process of research, as well as to those researched. social consthuci ionist \ini> inn HiMtinvi; aimmioac mi:s '•■>■ i.il constructionist and interpretive approaches are enormously varied. W'lial lliey share, however, is the notion that all social reality is constructed, .....eated, by social actors. These approaches ask us to focus on interaction: I Im\v do humans act toward one another and the objects in their worlds? W11.it meanings do they attach to them? Interpretive approaches in social research are closely related to a theo-m i u al tradition called symbolic interactionism, which rests on three premises I lilumer 1969, p. 2).The first is that humans act toward things based on the meanings those things have for them. For example, a European American mirjit look at a bundle of bamboo or wood sticks and think of them as kindling or as merely sticks, without a particular use or purpose. But someone II mi n China or someone who has traveled throughout Asia might look at ili.it same bundle of sticks and see pairs of chopsticks. And to my daughter I- .ii herine, that same bundle of sticks might become a group of imaginary lueiuls, or dolls, who are playing together. Each of us acts toward the sticks I .i ai ting a fire with them, using them to eat with, playing with them) on the I usis of the meanings they hold for us. The second premise is that the meanings of things arise out of social interaction. For example, in Chinese culture, the notion that bamboo sticks ne eating utensils called chopsticks (or, more accurately, kudi zi) is concluded through group life. The sticks have a special name that all can i '< ognize. Children learn to eat with chopsticks at a very early age, and ' liopsticks are offered in restaurants and households as appropriate eating utensils. If someone hands you a pair of chopsticks, it is understood that i liey are to be used to eat with, not to build a fire with or plant with or for ■.oine other use. The third premise is that meanings are created (and changed) through a process of interpretation. There is nothing in the bamboo or wood itself tlial tells us whether the sticks are dolls or eating utensils or fuel or any oilier thing. We understand their uses—that is, we create their meanings— through a process of interpretation. Thus, when chopsticks are placed beside a bowl at mealtime, we interpret that they are intended as eating utensils. While waiting for her food to arrive, however, my daughter may interpret them as toys. And if, in interaction with my daughter, I interpret I lie sticks as eating utensils, I may become annoyed and ask that she stop playing with them. But if I interpret them as toys, I can play along with her. because humans are social creatures, however, our interpretations of reality are not just individual creations. Although my daughter might prefer to see chopsticks as dolls or toys, hers is an idiosyncratic reading* ( i eating and interpreting reality are essentially social processes. What are the implications of this process of interpretation lor qualitative research? First, this interpretive tradition assumes that researchers need to begin by examining the empirical world. That is, rather than begin with a theory or preconceived notion of the way the world works, researchers should begin by immersing themselves in the world inhabited by those they wish to study. This initial approach is similar to that of the naturalistic perspective. But instead of seeking to go "inside the worlds of their subjects," the researchers' emphasis is on understanding how individuals construct and interpret social reality (Gubrium and Holstein 1997, p. 38). There is no social reality apart from how individuals construct it, and so the main research task is to interpret those constructions. The focus is on how given realities are produced. For interpretive scholars like Clifford Geertz, interpreting social reality is a lot like doing clinical work. Given a particular set of symptoms, these scholars ask, What could they mean? How can they be interpreted? Like a clinician, an interpretive researcher doesn't predict (as is the goal of the positivist researcher); rather, she or he diagnoses. How do you tell if a piece of interpretive research is good research? It must ring true, or at least seem plausible, to the participants themselves, and it must help to explain the "symptoms." Because researchers, too, are human, the research process itself can be seen as a social production. The meanings of research are negotiated between and among researchers and research subjects, as well as among other social researchers. Researchers begin with the constructions social actors use to define what happens to them, but they do not stop there. As Clifford Geertz (1973) reminds us, researchers never truly capture the viewpoints of others. Researchers' writings are always interpretations of what they think their research subjects are doing. But their insights are always limited, because they cannot know for certain what is really going on. In this regard, interpretive writing is akin to fiction, in that it is fashioned from a researcher's interpretation, or best guess, of what is going on. But it is not wholly fiction because it is rooted in social actors' actual lives; it is not simply made up.** Or, at least, it is partially so. I am indebted to an anonymous reviewer who suggested that the use or chopsticks as dolls isn't really idiosyncratic. Rather, it can be understood within the social construction of childhood as a distinct period of life organized in particular ways—for example, characterized by "play" instead of "work." **More recently, interpretive scholars have begun to examine the boundaries between fiction and social research. We will discuss these efforts shortly. Tlie next tradition we'll consider is feminism and critical research. In ...,„„. lespects, the division between various traditions is not clear-cut. There „, ..„„„. researchers, for example, who think of themselves as feminist and ,., unstnictionist. Nonetheless, the distinction helps highlight some impor-i mi differences in the aims of social research. i i MINIST AND OTHER CRITICAL APPROACHES < iitical Social Research feminist and critical researchers are a diverse group comprising many re-'.earchers who might not be happy to be lumped together in this way. Generally, critical social research, including feminist research, seeks insight into ilie social world in order to help people change oppressive conditions. In iliis context, criticism doesn't merely mean judging negatively; it also means, as feminist scholar Joyce Nielsen notes, exposing existing belief •.tinctures that "restrict or limit human freedom" (1990, p. 9). Whereas the goal of positivist research, described earlier, is to "predict and control" and I lie goal of interpretive research is to understand and interpret, the goal of t, i itical social research is to work toward human emancipation. For example, feminist researcher Pierrette Hondagneu-Sotelo (1996) studied immigrant women who did housecleaning in private households. But she wasn't interested only in how these women organized their work lives. She also wanted to find ways to act as an advocate to improve the women's working conditions, which typically included isolation, low pay, and opportunities for exploitation. With this in mind, she and her colleagues developed a set of novelas, or booklets that resemble comic books, that explain domestic workers' rights and some strategies for increasing their pay and decreasing the possibility of exploitation. Like Hondagneu-Sotelo, critical researchers pay close attention to the underlying mechanisms that account for unequal social relations. They want to examine the nature of inequality and work toward the empowerment of those with less power. Thus, they want to understand not only people's subjective feelings and experiences but also the material world and power relations within it. Because oppression is reproduced most easily when people view oppressive conditions as natural or inevitable (Kincheloe and McLaren 1998, p. 263), many critical researchers focus on how oppressive conditions are constructed and maintained over time. In doing so, many critical researchers hope to uncover myths that maintain oppression (for example, myths about wellare mothers or immigrant doinesli. winker,). Critical researchers also hope to communicate their findings to people especially the people they study—so that they can use them to light oppression. How can you tell if an explanation is true or false in critical social science? As a first step, it is important to know if the descriptions are plausible to those being researched. At the same time, "good" critical research teaches people about their own experiences, gives them insight into their place in the social world, and helps them transform the world. Because of the nature of oppression, those in less powerful positions may not always be able to see clearly the ways in which their reality is shaped and limited by what Dorothy Smith (1987) calls the "relations of ruling." Critical social science is action-oriented; thus, values are involved. Critical researchers argue that social research is, itself, a moral and political activity. Scientific activity is never neutral, and it can be used as a source of social control. Thus, researchers must not only be aware of their own values but also embrace a set of carefully considered values. Feminist Research A feminist approach in sociology emerged in the late 1960s and early 1970s out of an interaction between feminist activists and sociologists (Laslett and Thome 1997). Since that time, a vibrant body of research and theory and a network of scholars has developed. Feminist approaches to social science are extraordinarily diverse. Although some feminist scholars may take a more liberal approach and others a more radical one, what they share is a sense that social science as traditionally conducted does not fully take into account the presence of women in social life and the range of women's concerns. When feminist scholars looked at traditional sociological topics like work and occupations, and organizations, and at existing theory, they found that women's perspectives were not included. Apart from a few areas such as sociology of the family, women were essentially missing in sociological theory and research. Feminist scholars have been among the most important critics of traditional ways of doing social science. Not all feminist social research is qualitative, and not all feminist scholars agree with one another on issues of theory and method. Still, feminist critiques have played an important role in transforming social research methodologies. Understanding the distinction between methodology and method might be useful in highlighting the depth of the challenge feminist scholars P'.-.ed Mrihoil.s are (lie actual (oof. 01 (ei Imiques that scholars might use, • t. h .c < oikIik ling a survey or inteivievv. Mrlhodolofy rolcrs to the "theory hi.I analysis ol how research should proceed" (I larding 1987, p. 2). Feminist ■ In •, have called lor a transformation not so much in the concrete meth-■■•I. dial social scientists might use but in the methodologies. And this in if. ■. sense. Researchers from a variety of paradigms might use very similar i. ■ Iiniques when observing or asking questions. But how they think about ill. se techniques and how they analyze the evidence they have amassed inay dill'er radically. Some of the earliest feminist scholars challenged social scientists to in. hide women as subjects in their research. They also encouraged social i'-searchers to study the contributions of earlier women social scientists w hose work had been ignored or forgotten, like Harriet Martineau and lane A. K lams. Increasingly, feminist scholars have argued that the very theories mil methodologies social scientists have used are fundamentally flawed, •i holars such as Dorothy Smith and Patricia Hill Collins recommend a I nndamental reshaping of the social research process. In an early influential statement, for example, Ann Oakley (1981) argued that the pretense of ob-i<-< tivity in interviews actually prohibits a deeper understanding of women's lives. In her interview studies of new mothers, she found that sharing her own experiences of mothering facilitated a much richer understanding of i he women. Dorothy Smith has argued for a fundamental transformation of sociology, urging that we create a sociology for women. Such a sociology would begin from the standpoint of women and be rooted in women's experiences ol daily life. But it would not end there. The purpose of such a sociology would be to locate women's experiences in a broader network of ruling rela-lions, which includes a complicated set of social practices and social institutions such as government, the military, business, and the media (Smith 1987, p. 3). It would seek to understand how women's lives are shaped by forces outside of their control. Yet, in looking at these larger-level forces, such a sociology must not lose sight of women's real and subjective experiences. Smith argues, "The development of a feminist method in sociology has to go beyond our interviewing practices and our research relationships to explore methods of thinking that will organize our inquiry and write our sociological texts so as to preserve the presence of actual subjects while exploring and explicating the relations in which our everyday worlds are embedded" (1987, p. 111). Just as feminist scholarship challenged traditional methods, so has it come under challenge. In its earliest formulations, feminist scholarship often focused on tlie experiences ol White, middle i lass, him I heterosexual women.The resulting theories and accounts thus did imi ueiessaiily relied the experiences of women in all their diversity. Women ol color, lesbians, disabled women, and others have challenged feminist scholars to be more inclusive in their scholarship (see, for example, Cannon, Higginbotham, and Leung 1991). Recent scholarship focusing on the intersections of race, class, gender, and sexuality has expanded and transformed this earlier feminist research. What are the implications of feminist critiques of social research? Shu-lamit Reinharz (1992) argues that feminism is a perspective, not a method. In her survey of feminist methodologies, she identified a number of themes that characterize feminist research, including a critical stance toward traditional methodologies and theories, the goal of creating social change, the desire to represent human diversity, and the attempt to think about the relationship between researcher and those being researched. At the very least, feminist scholars argue, the experiences of women in all their diversity are important and must be included in designing and carrying out research. At their most expansive, feminist scholars argue that traditional methodologies themselves must be transformed. POSTMODERNISM The final research tradition we'll consider is postmodernism. Some scholars believe that social conditions have changed so dramatically that we have entered a new, postmodern era in which previous ways of knowing are no longer useful (see, for example, Gergen 1991). As traditional bases of authority have been undermined, they say, a variety of competing perspectives have replaced established ways of knowing. Although those who have been influenced by postmodernism may not agree on much, they tend to agree that there is not one reality, but instead a number of different realities and ways of knowing, all equally valid. Some argue that the postmodern world is, increasingly, a world made up of texts and images characterized by a hyperreality, "a term used to describe an information society socially saturated with ever-increasing forms of representation: filmic, photographic, electronic, and so on" (Kincheloe and McLaren 1998, p. 269). As people are exposed to this rapidly growing number and variety of images and types of information, there is the erosion of what some call "master narratives," or single theories as all-encompassing explanations. There can no longer be one coherent, objective theory i" i Nphin .mh ial phenomena, but lalhei multiple- stories, positions, and < ■ pii \eiilntions. Postmodernism poses a crisis lor previously accepted ways of knowing I mil, hence, for qualitative research) that centers on two issues (Denzin and I ni< oln 1998). first, postmodernists question whether "qualitative re- ■ aii hers can directly capture lived experience" (1998, p. 21). If there is no ..... objective reality, then a researcher cannot, of course, capture that real- ii v in a study. If the self is fragmented, as scholars like Kenneth Gergen I I'»') I) argue, then how can a researcher be fully reflexive (see also Gergen • iik I (iergen 2000)? How can a researcher maintain a unified stance toward i lie subjects of his or her research? The second issue revolves around what has been called the "legitimation ■ usis" (Lincoln and Denzin 2000). This crisis arose when anthropologists ■ind other social scientists criticized the authority of the written text—that is, the idea that texts can be considered "accurate, true, and complete" (Lin-« oln and Denzin 2000, p. 1051). In the postmodern perspective, texts are always partial, limited, and rooted within a particular viewpoint. If all knowledge is limited and socially constructed, then how do we evaluate and interpret social research? If there is no one reality, but merely a variety of perspectives, then traditional criteria for evaluating and interpreting social t > search simply are not relevant. Not surprisingly, scholars who have been influenced by postmodernism are not a unified group. There are substantial debates about what postmodernism is and what its implications are for doing qualitative research. For example, if there is no single reality or truth to be told, then there is no one "right" way of doing research or interpreting data. Rather, there are multiple stories, from multiple perspectives, that might be told. Qualitative research accounts are always incomplete and uncertain, because all knowledge is provisional. This has led some researchers to experiment with new forms of research—what George Marcus called "messy texts" (1998). So, for example, some researchers have experimented with the boundaries between qualitative research and fiction (Krieger 1991) and poetry (Richardson 1992). In her study of unmarried mothers, Laurel Richardson (1992) decided to write a poem, using the respondent's own words, to portray the life story of one of her informants, named Louisa May. She felt that by doing so she could portray Louisa May's life much more faithfully than if she had produced a simple transcript. In addition, if all knowledge is created by someone, then it is important to understand who that someone might be. Even if the self is, ultimately, fragmented and unknowable, it is important to consider the ways in which researchers are situated in particular, local settings. MAKING CLAIMS AllOUT I'AltADK .MS By now, you might he wondering why you've been reading what seems to be a philosophical discussion. Why should you care about paradigms? Whv does a book about research methods have to be phrased so abstractly? I argue that if you don't think about them explicitly you will still be operating within the constraints of a paradigm. You just won't be doing it consciously, as a result of choices that you make. Throughout the rest of the book, I will try to show you how the choices you make about research traditions influence the research strategies that are available to you. For now, 1 urge you to think about which tradition(s) seem to make the most sense to you and which ones seem most plausible. This is a first step toward creating your own research project. QUESTIONS FOR THOUGHT 1. Think about the different traditions you have just read about. Which ones seem more reasonable to you? Why? Do any seem less plausible to you? Why? 2. Have any readings in sociology particularly interested you? If so, which ones? What kinds of perspectives did they take? Why were the readings so compelling? 3. Think about who you are. What kind of family do you come from? Where do you live? What kinds of experiences have been most important to you in shaping your values? Now, think about someone who seems very different from you. If you are relatively young, you might think about someone who is older. If you live in a city, you might think about someone who lives on a farm. Or, if you are gay or lesbian, you might think about someone who is heterosexual. Now consider a social issue that you see as important. How might you go about researching it? How might someone who is very different from you think about the same subject? EXERCISES 1. Try this one with a friend. Go to a public place-a library, shopping mall, cafeteria, or similar place—to observe. Observe for about 20 minutes! While you are doing so, take careful notes on everything you think is uii|iuit;iii(. Alter you have finished, ie.nl ea< h other's notes. Did yon • •• -ill ihiIh e the same (lungs? What dillerenees do you see? Why do you • hmls you saw things differently? Mow did your personal experiences rive you somewhat dillerenl perspectives? I s< ■ i y day lor a week, read a national newspaper, such as the New York limes. As you go through the paper, look for stories that report the 11-.ults ol social research. How can you tell that social research is being n-porieil? Can you think of any ways that the research being reported might be useful? li y to hnd several qualitative research reports in the library. (You might browse, for example, through journals like Journal of Contemporary l.thnography, Gender 3d Society, Symbolic Interaction, or Qualitative Inquiry.) If your library provides access to a computer database that lets you print out whole articles, like Infotrac or EBSCOhost, you might search any topic that interests you. Or you can search using terms like "qualitative research" or "qualitative methods" or "qualitative study." These will yield too many citations for you to look at each one, but you can browse through to find ones that interest you. Alternatively, you might try to find writings by some of the researchers mentioned in this chapter. Once you've located several qualitative research reports, see if you can find descriptions of the researchers themselves. Do they give any personal information about themselves? Do they describe how they became interested in the topic? How do you think researchers' personal lives affect the choices they make? ■1. Go back to the research reports that you found in Exercise 3. What kind of reasoning did the researchers use—inductive or deductive? How can you tell? SUGGESTIONS FOR FURTHER READING Examples of Qualitative Research Becker, Howard; Blanche Geer; Everett Hughes; and Anselm Strauss. Boys in White: Student Culture in Medical School. New Brunswick, NJ: Transaction, [1961] 1977. A classic Chicago School study of the socialization of medical students. Burawoy, Michael, et al. Ethnography Unbound: Power and Resistance in the Modern Metropolis. Berkeley: University of California Press, 1991. A collection of short ethnographies written by graduate students at the University of California, Berkeley, each of which uses the process of theory reconstruction. Ki unit i, IV Ii im ic. C '.Hilling Selves: Jimiti; ( lender, itittl Duiminri Iileiihiy tu « i nese Workplace. Chicago: I Inivcrsity ol (lliuigo l'lcss, I 'i'M) An innovative study, influenced by postmodernism, of sell and identity in a lapancsr woikplacc. Liebow, Elliott. Tally's Corner. Boston: Little, Brown, l!)(i7. A (.lassie licld study of urban Black men. Stacey, Judith. Brave New Families: Stories of Domestic Upheaval in Late Twentieth Century America. New York: Basic Books, 1991. A feminist analysis of postmodern family life. Stack, Carol. All Our Kin: Strategies for Survival in a Black Community. New York: Harper, 1974. An important study of family and kinship in an urban Black community. Whyte, William Foote. Street Comer Society, 4th ed. Chicago: University of Chicago Press, [1943] 1993. A study of Italian Americans in Boston's North End; one of the most widely read and influential field studies. Resources on Research Paradigms Lincoln, Yvonna S., and Egon G. Guba. "Paradigmatic Controversies, Contradictions, and Emerging Confluences." Pp. 163-188 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage, 2000. An overview of research paradigms and the controversies surrounding them. De Vault, Marjorie L. Liberating Method: Feminism and Social Research. Philadelphia: Temple University Press, 1999. A look at research from a feminist perspective. General Resources on Qualitative Research Denzin, Norman K., and Yvonna S. Lincoln (eds.). Handbook of Qualitative Research, 2nd ed. Thousand Oaks, CA: Sage, 2000. A comprehensive guide. Strategies for beginning Research getting started: where to begin? < ietting started is often one of the hardest tasks for beginning researchers. I low do you know if you have a good idea for a research project? How can you tell if the research project is doable? Just as important, how can you tell if the research is worth doing? Even experienced researchers sometimes have trouble finding a topic, beginning a new research project, figuring out how to start, choosing a research strategy, or developing a general research plan. The first step in any research project is deciding on a general topic and then refining the topic. Suppose you are interested in homelessness or in students' alcohol use. Homelessness in general would be far too big a topic—even for someone doing a Ph.D. dissertation—so you would have to find a way to make your research more manageable. You could focus on just one aspect of homelessness, such as the experiences of workers in homeless shelters or the effects of homelessness on children. To narrow down the topic of student alcohol use, you could focus on the role of alcohol at student parties or the effects of alcohol consumption on academic achievement. You also need to think about the different approaches you could take to your research, as reflected in the paradigms discussed in Chapter 1. But what do you do if you don't know what general topic interests you? How do you begin then? 25 (il INIItAIINC; 11)1 AS How do you generate useful, interesting ideas? It's hard lo give sound advice on finding good research topics, for several reasons. One problem is that we don't usually think about where good ideas come from. They seem somehow serendipitous, a stroke of magnificent luck. Suppose you are walking down the street and suddenly notice that most of the streetlights are out on one block but functioning on the next. You also notice that the trash seems to be picked up on that block but not on the first one. This leads you to wonder about how different neighborhoods gain access to city services. Or suppose you read an article in the newspaper or see a program on TV that sparks your curiosity about how homeless people survive in the winter in cold climates. Or maybe you have a long-standing interest in sports or young children, and so research ideas seem to come naturally. These are all legitimate ways to identify a research topic. If you don't have any interests at all, you'll find it difficult to find an interesting research topic. But most of us are interested in at least something. Often, qualitative researchers begin where they are. That is, they look at their own lives to see if they can find anything interesting to study, an unusual angle or puzzling event or phenomenon. Then they try to refine the topic into a more manageable—and researchable—form. For example, Carol Freedman, a graduate student raising a young child, needed to do a research project for a course on research methods. She had been participating in a mothers' group, and so she decided to study it. The project eventually became her master's thesis, titled "Setting Stay-at-Home Standards: An Ethnographic Study of the Southland Mothers Association" (Freedman 1997). Elliot Liebow wrote in the preface to his book Tell Them Who I Am that he had been diagnosed with cancer and had a limited life expectancy, so he decided to volunteer at a soup kitchen. As he put it, "I did not want to spend my last months on the 12th floor of a government office building, so at 58 1 retired on disability from my job of 20-some years as an anthropologist with the National Institute of Mental Health" (1993, p. viiij. Because he felt pretty good for a lot longer than he expected, he started volunteering at a homeless shelter as well. He became interested in the lives of the women at the shelter, so he began taking field notes and thinking about the shelters as a site in which to do research. Ultimately, he did an in-depth study of the lives of homeless women. Other researchers, too, have written about how they developed their research interests. Lynn Davidman (1999) wrote about the experience of losing her mother to cancer when she was 13 and how that shaped her ■ I. ' imuii lo siiiily what she i alls "nioiliei Iiiss." The point is, il you look iuc you've determined that you tan gain tn iev. in the cniiip vou arc interested in studying, you need to think about vvliat otlni usources you will need. First, consider time. Doing qualitative resean h cm lake a great deal of time. If you are trying to study a group to which it may be difficult to gain access or that may be hostile to researchers, be sure you have enough time to develop the kinds of relationships you will need to do the research. For example, suppose you are interested in illegal drug use, as anthropologist Steven Koester is (Koester 1994). Specifically, you are interested in how HIV might be transmitted among street people who inject drugs. It will take a long time before they trust you enough to confide in you, so you need to consider whether you have enough time to gain trust. Another important resource is money. Doing qualitative research can cost money. If you are going to do the research full-time, you still need to support yourself If you are going to interview people, you need to purchase or have access to a reliable tape recorder and audiotapes. You also might have to hire someone to transcribe interview tapes, unless you plan to do it yourself. You may have to travel somewhere else to get to your research population or to find documents in an archive. You may need to purchase films or other texts to analyze. And there may be other costs as well. The question of feasibility can be particularly difficult if you are trying to conduct a research project over the course of a semester. You may have great ideas for research projects, but the projects are too ambitious to be carried out over 14 or so weeks. Once you finally gain access to your research population or data, it is the end of the semester and time to wrap things up. Thus, you may find it easier to begin with a setting that you already know or can gain access to. But this means you'll have to be especially careful about the preconceptions and biases you bring to your work. Is the Research Worth Doing? A final issue is whether you should do a particular project. Just because you are interested and have the resources you need, it doesn't mean you should actually do the research. The key issue is whether the research has any potential uses or benefits. Will your research make a contribution, either to individuals or to a larger group or to our general knowledge base? Does it have the capacity to harm anyone—either yourself or the research participants? Before you begin any research project, you need to consider the potential benefits and risks. (These ethical issues will be considered in detail in Chapter 3.) Again, let's say you are interested in researching people in homeless shelters. You think the study might have potential benefits because people .ni'lit In- more sympathetic to the piuUems ol (he homeless il they under-.i...M| what their lives were like. They might be more interested in building illonlable housing or having the government spend more money on subsi-li-eil homes. While the research might not have an immediate positive ■ i >11 >:n I on the participants' themselves, you think that in the long run your •soi Is might help debunk stereotypes about homeless people. I)l VI LOPING A RESEARCH STRATEGY < imc you've chosen a topic and framed a research question, you need to i levelop a research strategy. Specifically, you need to address these questions: «• How will you gather the data? * What kind of population or setting will you study? » Will you use in-depth interviews, or do an observational study or work with "texts" (which can include things like books and magazines but also media such as TV shows, movies, and songs)? » How will you begin to analyze and make sense of the data you have collected? I different research traditions suggest somewhat different strategies. We'll consider these next. Different Traditions, Different Starting Points Depending on which research tradition you choose, you will begin your research from very different starting points. Before you get too far in your work, then, you need to consider which tradition(s) makes the most sense to you. According to the positivist paradigm, which we discussed in Chapter 1, the goal of social research is theory testing. Thus, in this tradition, you need to settle on a theory before you begin your research. Researchers who work within this tradition usually spend much time researching what others have found about their topic. They then develop hypotheses, or statements that can be tested, based on these theories, which often are framed in causal language: "x causes (or affects) y" or "The more of x, the less of y." Then they develop a research design that they can use to test their hypotheses. They use the results of their empirical tests to determine whether their theory is useful. Let's say you're interested in rational i lioi\ <• iln-niy (Ineilinan ami I lechler 1988), which stales that people act act online, ln theii bfsi inter ests. Specifically, you're interested in applying rational i hoiYo theory in divorce. You think that, if people feel they will get more out of divorcing than staying in a marriage, they will choose to divorce. You're aware that raising children after a divorce can be hard, so you think that people with children have less to gain from a divorce than those without. Your hypothesis might be this: Couples who have children are less likely to divorce than couples who have no children. You could test this by comparing the divorce rate of couples who have children with that of couples who do not. Qualitative researchers rarely work within this positivist tradition. That is, they are much less likely to test hypotheses than are quantitative researchers and are much more likely to work within one of the other traditions discussed in Chapter 1. Instead of beginning with a theory, qualitative researchers are more likely to begin with an examination of the empirical world. In the naturalistic and constructionist traditions, researchers immerse themselves in the social worlds of their research subjects. Only when they have been in a setting for a long time do they begin to develop theories. Some call this a grounded theory perspective (Charmaz 2000; Strauss and Corbin 1998), in which the aim is to develop theory grounded in the empirical world. If you choose this approach, your first step is to decide on a field setting or site for your research. At this stage, although you should do a library search to determine if others have studied the same kinds of sites, you should not try to develop testable hypotheses. Rather, you should focus on how you might gain access to the site and begin building relationships with the people there. Researchers working within a critical research tradition might decide to do action research, in which the objective is to produce some kind of social change. For example, you might work with a coalition that seeks to end homelessness in your community. In this case, you first need to identify the stakeholders—the people who have a stake in eliminating homelessness. Obviously, people without homes do, but who else? Shelter workers? Community agencies? Neighborhood residents? You also need to identify who has the power to effect change and what the people you are working with think needs researching. For example, they might believe that the research should focus on the cost of housing in the community and on people's wages. In action research, rather than begin with a theory, you need to begin with a clear statement of the needs and priorities of the group. While the discussion of research paradigms in Chapter 1 may have seemed abstract, the choice of a paradigm has real-life implications. The . Iioit e\ you eventually make will determine whether you turn first to the i 'ill. I r, I led i esean h, or to the em pi i k a I world. In research that draws on the li-.itivisl tradition, you need to have well-developed hypotheses before you !■• r.m your research. In research that draws on the naturalistic or construe-n..iii',i traditions, you need to frame a general research question and choose i lie lor your research. In research that draws on postmodernist traditions, ■.on might locus on texts. In this approach, although it's a good idea to read 11 least some of the published literature before you begin (and many n .eaivhers, myself included, would insist that you do so), you don't begin u iih already formed hypotheses. Instead, you develop your theory and an mi icasingly refined research question as you conduct the research. I\|)cs of Research Strategies ' hue you have settled on a research question and research tradition, you need to decide how you will collect your data. Specifically, you need to address these issues: « What research strategy will you use? * What population or site will you study? » What texts will you choose? * What will your evidence consist of? Transcripts of interviews? Observational notes? Archival materials, like letters or diaries or an organization's records? Songs or videotapes of TV programs? * How will you spend your time? Listening and talking to people? Observing? Going through published materials? Watching audiotapes? You can choose from several general research strategies. Which one you settle on depends on your research question, the research tradition you see yourself as working in, and your own individual preferences. Observational Studies In the naturalistic or constructionist traditions, you might conduct an observational study, in which you gather data by observing interaction in a particular site (such as a street comer or homeless shelter). Observational studies are useful when you want to understand how people behave in a particular setting or when you want an in-depth understanding of a particular culture or group. In an observational study, you might choose to participate, lor example, you might volunteer at a homeless shelter, as Elliot Liebow, whom we discussed in Chapter I, did. t >i ymi might choose simply to observe in a public place (such as a shopping mall or a public park), without participating. Interviews Many qualitative researchers choose to conduct formal in-depth interviews with people. These can be relatively structured or unstructured. Interviews are good research techniques when you want to know what people think or feel about something. Researchers often combine observational techniques with either formal or informal interviews. In formal interviews, the researcher sets a particular time and place for an interview. Informal interviews tend to arise spontaneously in the course of observation. For example, you might decide to formally interview people who volunteer in a homeless shelter in order to understand their experiences, as well as to observe and informally interview shelter residents. Unobtrusive Measures Not all research involves talking with or observing people. Unobtrusive research involves examining human traces, or evidence of human activity. For example, if you want to know which magazines in the library are most popular, you might study which ones seem to have thumbed pages or seem to have been heavily perused. A number of researchers have studied public graffiti. Caroline Cole (1991), for example, analyzed the writing on walls in women's bathrooms, arguing that the graffiti there served as an alternative means of communication. Jeff Ferrell (1995) analyzed hip-hop graffiti in Denver, Colorado, combining participant observation of graffiti writers with visits to graffiti sites in other cities. He argues that hip-hop graffiti reflects young people's efforts to resist social control. Sometimes, researchers study "texts" such as newspapers, books, organizational records, TV shows, and court transcripts. For example, Sharon Hays wanted to investigate what she called the "cultural contradictions of motherhood" (1996). To do so, she analyzed child-rearing manuals to identify the kinds of social norms for mothering contained in them and conducted in-depth interviews with mothers of small children to determine how they actually viewed their mothering. Triangulation Each research strategy has particular strengths and weaknesses. For example, in-depth interviews can provide insight into people's thoughts and feelings, but people's behaviors don't always match their words. Analysis of texts can tell you about social ideals for behavior, but the texts can't tell you how people actually respond to them. For this reason, ,e-,eaiehers often ust' two or mote research strategies.This is called triangu-l.uioii. because different data collection strategics have different strengths and weaknesses, research designs that include multiple research strategies tend to he the strongest ones. It LADING THE LITERATURE Whichever research tradition and strategy you choose, you should visit the library early in the process of designing your research. Although some natu-i alistic researchers caution against becoming too wedded to a particular theory or viewpoint before immersing yourself in your field setting, I think this concern is a little overstated. By knowing what other researchers have already said about your topic, you are in a better position to come up with a well-thought-out research plan. And at some point during the research process, you will still need to conduct a literature review to help you place your own research in context. I recommend that you begin any research project by simply browsing. I ,ook through the journals and books that seem most interesting to you, or browse through the databases available at your library. I usually skim through the abstract or the introduction quickly to see if I might want to read the whole article or book. Then I go through those readings that seem most useful in more detail. I recommend looking at as wide a variety of sources as you can. As you do so, be sure to take good notes (including accurate citations) so that you can locate the sources again as needed. There's nothing more frustrating than knowing that you had the perfect source but being unable to use it because you can't find it again. Every library is different. Some have subscriptions to many journals and excellent on-line searching capabilities; others don't. Each library has its own special way of providing access to the materials. Some libraries have on-line search services, like EBSCOhost or Infotrac, which will deliver whole articles to you on-line. Others have a large selection of books and journals in print that you can browse through. Your librarian or professor will be the best person to help you search in your own library. Most libraries will have access to Sociological Abstracts, either in print or on-line. One of the most important databases for sociologists, Sociological Abstracts summarizes the articles in the most important journals that sociologists publish in. By searching this index, you should be able to get a good idea of what others have said about your topic. Depending on your Iirlil and topic, you may also want to s<'ari li databases sn< h as I'syt hologii nl Abstracts, ERIC, or Criminal Justice Abstracts. Again, I shonglv leeomniend that you check with your professor or librarian to see what resources on your campus might be helpful for you. Although this book can't help you search your specific library, it can give you some general tips on conducting a useful library search: 1. Try a number of different terms for the same thing. Different databases will often use somewhat different key words, and the same database may yield very different articles if you use just slightly different search terms. 2. If the term you've searched yields too many citations to look through, try narrowing it down. For example, I recently used the on-line version of Sociological Abstracts to search the term "working women." When I did so, I received 3669 "hits"—clearly, too many to look through. When I narrowed the topic to "working women and sexual harassment," I received 49 citations—still a lot to look through, but a much more manageable number. 3. Try a number of different sources, including the book catalogue, the journals (also called serials), and the Internet. Different sources will tend to give you very different kinds of results. 4. Once you've found a useful article or book, see if you can track down some of the sources the author used. They are often helpful. 5. Ask a librarian for help, especially if you're finding very little information about what you think should be a popular topic. But be sure you've already thought about some potential angles for your project. The librarian can't create a research topic for you, but she or he can help you find the right resources, given a specific topic. EVALUATING WEB SITES The Internet can provide a wealth of material for research projects. You can get information from a variety of government and private sources, as well as research reports, book reviews, and other useful texts. Many government agencies have Web sites, including the Census Bureau, the Bureau of Labor Statistics, and the FBI. However, the Internet can also lead you astray. A friend of mine says that the World Wide Web is like a huge catalogue— everyone wants to sell you something. Thus, you need to evaluate Web sites m 11 ill v helm e i el y ill); on I lie in in you I i eseai i li. I I ere are a lew questions i.l' in judging the uselulness ol vai ions sites: Who sponsored the site? Is it maintained by a commercial enterprise or individuals who stand to gain something? Does it contain advertising? Is it maintained by a government agency or university or research institution? Or is it maintained by an individual? If so, what qualifications does the person have? Is she or he an expert in the Held, or someone who is mainly trying to express a personal point of view? Does the site seem obviously biased? Does it use obviously inflammatory language? Is it published by a political or social organization with a particular agenda? Does it have a mission statement or statement of purpose anywhere on the site? What kinds of sites is it linked to? - How often is the site updated? If it was published several years ago and hasn't been updated since, it probably is not a particularly reliable source. As you continue to develop and refine your research project, you will probably need to return to the library a number of times. As your research i hanges, so will the literature that you find useful. But these tips at least will get you started. THE RESEARCH PROCESS IN REVIEW I developing a good research project is an ongoing process. There are a number of steps you need to take when embarking on a research project. Although the steps are listed in one order here, you will probably find that you need to go back and forth between the steps as your research unfolds. You should also realize that your project may shift its focus as you learn more and gather more evidence about your topic. At each of these stages, try to remain open to alternative paths. 1. Choose a general topic and try to refine it into a research question. 2. Evaluate whether your topic is interesting, feasible, and worth doing—and ethical. Chapter 3 discusses ethical issues that may arise in research. 3. Develop a research strategy. Decide on the kind', ul d.il.i \'"U will collect, and think about how you will try to analyze m make sense of them. Make sure that the strategy is consistent with the research question you pose and the research tradition you are working within. 4. Begin your search of the literature. Although you will probably need to return to the literature at other stages in the research process, you should begin with a good sense of what others who have studied the same topic have found. 5. Begin collecting and organizing your data. Chapters 4-7 examine different methods of collecting data, including observation, interviews, textual analysis, and action research. 6. Begin analyzing (or interpreting) your data. Although qualitative researchers usually move back and forth between analyzing and collecting data, it is sometimes helpful to think of the two as separate steps, at least initially. Chapters 8 and 9 discuss strategies for interpreting data. 7. Write up your research. In qualitative research, it can sometimes seem as if you are writing all the time. Certainly, the process of organizing data entails writing, as does interpretation. Qualitative researchers present their work in many venues: as journal articles or books, as presentations at professional conferences, at training sessions or in-service meetings for professionals, and as presentations for community groups. Chapter ] 0 focuses on strategies for writing up your research and presenting it to a larger community. QUESTION FOR THOUGHT Now is the time to consider where you stand in relation to your research question. Think back to the research traditions, or paradigms, outlined in Chapter 1. Which one(s) seem most convincing to you? Do you think researchers should take a critical position? Or does a more traditional orientation appeal to you? What do you think are the strengths and weaknesses of the various approaches? What personal insights do you have into your proposed research topic? What kinds of special knowledge might you have by virtue of your own life experiences? How might these insights shape the kinds of research you might do? EXERCISES I As yon go about your daily routine, try to pay particular attention to your surroundings. See if you can develop at least two research questions sparked by what you encounter in your daily life. .'. evaluate one of your research questions in terms of whether it is interesting, feasible, and worth doing. Consider what kinds of resources you would need to complete the research. \. Think of a general research topic that you might find interesting. Can you come up with three or four different perspectives on it? Try refining the topic into several different research questions. I. Go to the library to see what information you can find about the research questions you developed. Locate at least five books or journal articles that you think might be useful. 5. Search the Internet to find at least two different Web sites of interest to you. Evaluate each Web site you've chosen in terms of its potential usefulness or bias. (J. Use several different research strategies to locate information about your topic. Try several different search engines on the Internet, such as Yahoo or Excite, and several academic databases, such as the card catalogue and Sociological Abstracts. (Check with your professor or librarian about the best methods for searching on your campus.) What different kinds of materials do you find using each strategy? SUGGESTIONS FOR FURTHER READING Creswell, John W. Research Design: Qualitative and Quantitative Approaches. Thousand Oaks, CA: Sage, 1994. A good introduction to research design, especially for those interested in comparing qualitative and quantitative approaches. Glassner, Barry, and Rosanna Hertz (eds.). Qualitative Sociology as Everyday Life. Thousand Oaks, CA: Sage, 1999. A collection of essays by qualitative scholars that examine the linkages between sociological research and everyday life. Lareau, Annette, and Jeffrey Shultz (eds.). Journeys Through Ethnography: Realistic Accounts of Fieldwork. Boulder, CO: Westview Press, 1996. A collection of researchers' accounts of doing qualitative research. LeCompte, Margaret D., and Jean J. Schensul. Designing and Conducting Ethnographic Research. Walnut Creek, CA: AltaMira Press, 1999. Part of the Ethnographer's Toolkit, a seven-book guide to conducting ethnographic research. 3 r.thical Issues I magine that you have been researching radical environmentalists—those .11 livists, like members of Earth First!, who believe that to achieve their y.iy.As they must sometimes break the law. You are interested in studying I low these groups operate and how their members are like or unlike participants in other social movements. You have amassed hundreds of pages of < oulidential notes, transcripts of personal interviews, and other materials. Now imagine that a federal grand jury wants access to your confidential materials for a case it is developing against some of your informants. On the < me hand, you guaranteed your research subjects confidentiality. They probably would not have agreed to talk with you otherwise, and they certainly didn't expect that you would be telling the police what they told you. On the other hand, you are being threatened with jail time yourself if you do not hand over the materials. What should you do? The chances that you will be threatened with jail time or otherwise run afoul of the law while conducting research are fairly slim. But this is exactly the dilemma that sociologist Rik Scarce encountered in 1993 while studying the radical environmental movement (Scarce 1995]. Scarce believed that his commitment to an ethical code meant maintaining his informants' confidentiality at all costs. Then a graduate student at Washington State I Iniversity, Scarce was jailed for 159 days for refusing to share his confidential information with the authorities. 43 SI II l»IN(i WEI,I, AT IN1C1IT: A PRACTICAL ETHICS Researchers who study controversial groups are not the only ones who face ethical dilemmas. Because social research is conducted by, for, and about people, there is always the potential to harm others. For this reason, all social researchers need to think about ethical issues. We need to consider a number of questions, such as these: 4 How should we conduct research so as not to hurt others? ♦ What kinds of relationships should we attempt to create with our research subjects? ♦ What kinds of power relations are there between those who are doing the research and those who are being researched? ♦ Who benefits from social research? Who should benefit? In my own current research on at-home mothers, it's very unlikely that a grand jury will try to force me to reveal my informants' names. Yet I still worry about ethical issues. I worry about whether I have treated those I am researching well and whether they will benefit from my research. I worry about what happens when I develop friendships in the process of conducting research and about whether I am interpreting the women's lives in ways they might disapprove of. At a very practical level, I like to sleep well at night. To do so, I need to know that I have treated my research subjects ethically. But what is ethical treatment of research subjects? How do I know if I have conducted myself in an ethical way? Researchers answer these questions in a number of ways. There is no one "guaranteed" way to act ethically in a given research situation, and reasonable people—all of whom may think carefully about ethics—may disagree. My aim in this chapter is to encourage you to think about the kinds of ethical dilemmas that research can pose. I hope you'll consider how your own belief systems shape your responses to important ethical questions. Ethical Codes To help researchers make ethical decisions, most professional organizations to which researchers belong (such as the American Sociological Association [ASA] and the American Psychological Association] have ethical codes. Codes of ethics provide standards for behavior. The ASA, for example, has a detailed code of ethics that "sets forth the principles and ethical standards ili.il underlie sociologists' piolessional i esponsibilitios ami conduct" (ASA i .«le ol Tlliics, p. I ). ( You can read (lie complete version of the ASA code of • ilixs on its Weh page: liltp://www.asanet.org. Parts of it are also reprinted pemlix A.) Although much of the code is not necessarily relevant for App I. -ginning soci al researchers, it also contains many useful guidelines. < onl'ulentiality and Informed Consent I'wo issues are of particular relevance for beginning researchers: (1) main-i.nning confidentiality and (2) obtaining informed consent. That is, researchers iinisl ensure that participants freely agree to participate in the research, and 11 iey must protect the privacy of their research participants. They also must inform participants of all potential risks from participating in the research ind gain formal consent before beginning. Although, as we'll see, the notion of informed consent may be especially tricky in qualitative research, there ue important reasons for considering the issue carefully before beginning. Concern for treating research participants ethically arose out of people's horror at the practices of Nazi medical researchers during World War II For example, they injected Jewish prisoners with spotted fever virus to keep the virus alive, and they deliberately infected other prisoners with .irep, tetanus, and gangrene. Some prisoners were placed in tanks of freezing water for up to 3 hours or forced to stand outside, naked, in below-freezing temperatures (Neutens and Rubinson 1997, p. 16). In order to stop these kinds of abuses, judges at the Nazi war crimes tribunal developed the Nuremberg Code in 1947 (Shuster 1998). The code provided basic standards for conducting research on human subjects. First, and perhaps foremost, research subjects must voluntarily consent to the 11 'search. Second, the research must show clear benefits to society. Third, the lesearch must be designed so as to minimize the potential of harm. Although the code contains other standards (for example, researchers must be qualified to conduct their work), the notion of informed consent has perhaps most influenced contemporary social scientists. Nazi scientists have not been the only ones to engage in unethical prac-tices—and, indeed, debates about the ethics of medical research continue today. Much closer to home is the Tuskegee Study funded by the U.S. Public Health Service. In this study, poor Black men who had been exposed to syphilis were left untreated for many years—even after penicillin had been developed and made widely available (Caplan 1992). About 600 Black men from Tuskegee, Alabama, were originally enrolled in the study. Promised free medical treatment (for diseases other than syphilis), lunches, and transportation, the men (at least, as one writer notes, those who were still living) were billowed Irom 1932 to 1973 in order to detei mine w li.il would hap pen il the disease were left untreated. (Syphilis is highly i ont.igious anil ian lead to serious heart problems, blindness, paralysis, insanity, and, ultimately, death.) It wasn't until a Public Health Service officer named Peter Buxtun aired his ethical concerns to the press that the study was stopped. Outrage over the Tuskegee Study led Congress in 1974 to create the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (Caplan 1992). This commission played a major role in the development of institutional review boards, or committees set up to consider ethical issues in research. Clearly, medical research has the capacity to harm—or help—its research subjects in direct ways. But what about social research? Can it be considered in the same light? Some people argue that it can and that social researchers must follow strictly the same kinds of guidelines as medical researchers. Other researchers argue that there are crucial differences (Morse 1998). Two influential research projects have highlighted some of the ethical issues in social research. In each case, the ethical issues are complex, and social scientists have disagreed—sometimes heatedly—about what the researchers should have done. CONTROVERSIES IN SOCIAL RESEARCH Milgram's "Obedience" Study One controversial study was conducted by psychologist Stanley Milgram in the 1950s (Milgram 1974). Milgram wanted to investigate how people are influenced by authority. He brought his research subjects into the laboratory and asked them to read pairs of words to another "subject" (really a confederate of the experimenter), who was sitting in another room, strapped to what looked like an electric shock machine. The research subjects sat in front of an elaborate control panel, with dials indicating voltages ranging from 15 to 315 and labels such as "extreme-intensity shock." The subjects believed that the object of the experiment was to see if the other person (the confederate) could remember the word pairs correctly. If the confederate answered incorrectly, the subjects were instructed by a researcher in a white lab coat to administer a shock. But the subjects couldn't actually see the confederate; they only saw the lights of the control panel indicating the answers and heard the responses to the shocks through an open door. As the experiment continued, the subjects were instructed to give what they thought were more and more intense shocks. When the subjects hesitated or H'lused, lite researcher ordi'icd them to lonlinue. As the shocks became mote intense, the conlederale su earned in (leigned) pain, kicked on the 11'ii a, and, dually, lell silent. When subjects (inally refused to give additional .lnu ks, despite the urgings of the researcher, the trial was over. What Milgram was interested in was when people would stop adminis-i' i in); the shocks.That is, once they felt they were hurting someone, would i hey refuse—even though a scientist, someone in a position of authority, pi essed them to continue? Many of the subjects did continue, experiencing I'.nal stress in the process. Some begged the scientist to let them stop; i it hers experienced physical symptoms of stress. Even though the subjects were not, themselves, subjected to electric shocks, many felt as if they had been. This study highlights several ethical dilemmas. Should researchers be allowed to deceive their subjects? Under what conditions? Once Milgram knew that the subjects were experiencing stress, should he have stopped the experiment? Could he have obtained the information he wanted in any i it her way? These are just a few of the ethical questions raised in relation to Milgram's study—and there are no easy answers. Qualitative researchers don't typically conduct experiments like Mil-gram's. Nonetheless, issues of deception can still come up—especially in types of research, like participant observation or in-depth interviews, in which researchers need to develop relatively close and trusting relationships with their research subjects. For example, should you tell your research subjects exactly what you plan to study? What if your findings are not necessarily flattering to your subjects? Or what if you are studying a group whose ideals you are critical of? Should you only study groups you approve of? For example, Kathleen Blee (1996) studied White women activists in racist groups such as the Ku Klux Klan and White supremacist groups. Blee describes herself as "decidedly unsympathetic" to the goals of the groups and, indeed, considers how her work could be used to help fight racism. Should she have informed her research participants of this before she began? If some of her research findings would help further the group's racist goals, should she share them? What should a researcher be obligated to tell his or her informants about the nature of the research? Humphreys' "Tearoom" Study A highly controversial study involving deception was conducted by Laud Humphreys in the 1960s and published as a book, Tearoom Trade (Humphreys 1970). Humphreys wanted to study homosexuality at a time when homosexuality was highly stigmatized (and illegal in most states) and when there was very little research about it. Specilicnlly, he was inteiesled in (hose men who had sexual encounters with other men in puhlu bathrooms, or "tearooms." If he had tried to pass out questionnaires or even simply oh serve, the men likely would have assumed he was a police officer and left the scene. Given the stigma (and legal consequences) attached to homosexual behavior, he probably would not have been able to conduct his study if he had approached participants directly. So, Humphreys decided to participate in the scene by acting as a "watch queen," someone who kept watch for the police so that other men could engage in sex. Humphreys did not re veil I his role as researcher; nor did he engage in sexual activities himself during the course of the study. But he secretly wrote down the license plate numbers of the men he observed. Later, under the guise of conducting another study, he was able to get the addresses of those men through a contact at the Department of Motor Vehicles. He then followed up with an in-person interview with the men, some of whom had wives and children, at their homes. He never revealed to the men how he had obtained their names and addresses or where they had encountered each other earlier. Again, this study raises ethical issues. Should Humphreys have acted as a participant in an illegal activity? Should he have told the men about his research? Could he have obtained the information any other way? Should he have followed up with the interviews in the way he did? Did the benefits of his research outweigh the risks? These are just a few of the ethical questions about his study—and, again, there are no easy answers. FEMINIST APPROACHES TO RESEARCH ETHICS Many feminist scholars have criticized researchers for not paying enough attention to ethical issues in conducting research. Although there is no sin-. gle feminist approach to research ethics, many feminist scholars believe that researchers need to address the power relationships that are embedded in research. Researchers, they note, often tend to be of a higher social class than the research participants. Researchers benefit from the research they conduct by publishing their findings and gaining public recognition, which helps them advance in their jobs. Researchers also determine how the research is conducted. In interviewing and other forms of data collection, they usually set the agenda and determine what is important, and they have the power to end the research relationship when they're finished. And it's their interpretations and analyses that get written up. The research subjects, while they may have the power to stop participating, do not have the power to leave the field that researchers do. Research participants can lie (and have Id d) in exaggerate their stories, but they do not typically have the power to ■ li iii mine, ultimately, how the data are used. I ■ ■■ 11iilist scholars argue that researchers need to consider these power i .ties when doing research. Most feminist scholars agree that feminist re-. in h should contribute in some way, even if indirectly, to social change 11'. Miliar/. 1992). Many have urged the development of close, personal ties I..i ween researchers and subjects (Oakley 1981), although others have H rued that such ties lead to a greater potential for manipulation and ethical hi. innias than other, more distant forms of research (Stacey 1996). Many I.. In ve that reciprocity can help reduce some of the power differences be-i >Min researcher and researched. That is, researchers may be able to offer dungs that are of use to those being researched, such as giving rides or tutor-mi' oi doing other favors that place researcher and researched in a more i.. ipiocal role. Still other feminist scholars believe that research should be .....ie collaborative in order to level, as much as possible, the differences in |...wer and status (Cancian 1996; Naples and Clark 1996). Many feminist . In .firs argue that researchers should share their findings with their sub-|. ■ is, and some believe that researchers should seek participants' responses nil I gel approval from them before publishing their work. We will consider i h. se issues further in Chapter 7, on action research. INSTITUTIONAL REVIEW BOARDS Mi1.1 colleges and universities—and other institutions that do research on humans—have review boards to make certain that researchers follow ethi-. .ii guidelines. These boards, called institutional review boards (IRBs) (or ..ineiimes human subjects committees), review research proposals before ill. i esc'arch begins to ensure that certain standards are followed. Each insti-luiioii has its own special procedures, but all must follow federal guidelines. 11h example, all IRBs must include at least one member who is not a social .....ntisl (for example, a lawyer or member of the clergy), and all IRBs must i nsiiiv that researchers have adequate procedures for obtaining informed i unsenl. Sometimes, research that is conducted under the supervision of an iii'.inictor as part of a class is considered exempt; that is, it does not need spe. ial 1RB approval. Other schools have special procedures for student ii '.catchers to gain approval. If you are planning to conduct research as part nl v< >ur program or class, you should check with your professor to see if you in id 1KB approval before you begin. I U Hs are unpopular with many researchers. You might wonder about litis. Why would anyone object to making researchers think about ethical issues? Who could not want research participants to he denied fairly? 'The problem is that many IRBs are more familiar with hlomedu al ami behavioral research that is experimental in design (like Milgrain's which probably wouldn't be able to get IRB approval today). Many IRBs are not as familiar with qualitative research designs and the specific kinds of questions that arise. For example, much qualitative research is open-ended in nature. Unlike survey research, in-depth interviews are not wholly scripted. Thus, it may be impossible for a researcher to tell an IRB exactly the questions he or she plans to ask research subjects. Because of perceived (and sometimes actual) difficulties in receiving IRB approval, some researchers have shied away from controversial topics or innovative research designs. ETHICAL CONCERNS IN QUALITATIVE RESEARCH Some Problems of Confidentiality and Intimacy Several features of qualitative research raise difficult ethical issues. The goal of qualitative research is typically an in-depth, detailed study. Instead of making statistical generalizations about a large number of cases, the goal is often to tell detailed stories (what some call "thick description") about a particular case or a small number of cases. And rather than use impersonal techniques of data collection, like surveys, which lend themselves to anonymity, the data are often collected after long periods (sometimes even years) in the field. Field researchers often develop close, intimate relationships with their research participants. Most qualitative researchers stress that the quality of the relationship between researcher and subject affects the quality of the research. Even when qualitative researchers conduct a one-time interview, it is still typically more intimate and personal. In a face-to-face interview, true anonymity is impossible. The researcher sees the interviewee in person, up close. Under these conditions, it is often difficult to maintain confidentiality. Imagine that you are studying people in a small community. Chances are, providing only a few details about a participant—like job title, race, and age—might be enough information for others to guess the person's identity In my first book, for example, I studied a small lesbian and bisexual community. One of the women I studied had a very unusual story. We both knew, when I was conducting the interview, that if anyone guessed the town in which the study was set, they would be able to tell who she was, even if I used a pseudonym (which I did) and changed identifying details (such as her occupation) in the book. We talked about the risks in the interview. In tins i use, she fell that her story was important, and she wanted other women to hear it. But what if she hail fell otherwise? Would 1 have been obligated to leave it out ol the book? Judith Stacey ('199(f) has written about some of the ethical dilemmas qualitative researchers face when they develop close, personal relationships with their research participants over a long period of time. While the participants (and the researcher as well) may see the relationship as one of friendship, the researcher still gains from that relationship. At some level, this iclationship can be exploitative. At the same time, some argue that those being researched may also benefit from the relationship. They may enjoy the opportunity to talk about themselves or like the fact that someone is interested in their lives. Indeed, Stacey acknowledged that her own research par-t icipants benefited in certain ways. Although there may be other, more tangible benefits to participation, however, researchers typically have much more to gain. It is, thus, a lopsided relationship, one that researchers can end simply by leaving the field. In addition, when participants experience hardship or personal troubles, researchers are in an ambiguous role. Should they act as a friend? Should they observe and take notes? What does it mean when a "friend" takes notes about another's hardship and then benefits from it? When bad things happen to research participants, it may simply be, to researchers, "grist for the ethnographic mill," as Stacey (1996, p. 92) puts it. Ethnographic researchers may find themselves concealing information or telling half-truths and otherwise behaving in ways that leave them feeling very uncomfortable—and in ways that might disturb research participants if they knew. Thus, researchers must be aware of the contradictions and tensions involved in doing feminist and ethnographic research. Some Problems of Informed Consent Because of the open-ended nature of much qualitative research, it is also difficult to provide truly informed consent. In a one-time interview, it is not too difficult to anticipate the risks of participation and to gain informed consent before beginning. But what if you are observing an organization for a long period of time? You might gain consent at the beginning of the study, before you begin. But if new members join several months after you begin, should you continue to announce your research objectives? Also, people might forget that you are a researcher. If your informants come to see you as a friend, they might feel subtly coerced into participating. As one researcher has commented, "A major part of the problem is that informed consent changes almost as frequently as the weather" ((!ov I'l'l'l, p ,') In (his type of situation, how do you negotiate and renegotiate inhumed ( onsent? You must also consider the types of settings in whkh you need to request informed consent. For example, you certainly need to gain consent in an interview or private setting in which all the members can be identified. But suppose you are observing a large public rally or demonstration—one with hundreds or maybe even thousands of participants. Should you attempt to gain consent from all the people there? Or suppose you arc observing a private meeting with a large number of people. What steps can you take to inform people of your research? Should you make any distinctions between private and public settings? That is, do people have a greater right not to be researched in private settings than in public ones? In addition to protecting confidentiality, researchers also need to be aware of some other privacy issues. At what point does social research invade the research participants' privacy? Because there are not always preestablished boundaries when conducting qualitative research, it's sometimes hard to tell when to stop collecting data. The line between collecting information for "legitimate" research purposes and invading respondents' privacy is a thin one. For example, suppose people share deeply personal information with you that, at least as far as you can tell, seems unrelated to your study. Should you encourage them to continue? Should you try to subtly change the subject? Or suppose people tell you intimate details about other people's lives—maybe people you know or other people involved in your study? What are your ethical obligations to them? Overt Versus Covert Observation Is it ever ethical to carry out research covertly—that is, without the participants' knowledge? Some researchers believe that deception is sometimes necessary for the research to take place at all. Humphreys' "tearoom" study would never have been done without it, for example. Others argue that deception is rarely or never ethical (see, for example, James Richardson's 1991 discussion of covert research in new religions). Richardson argues that covert research isn't really necessary most of the time—and probably isn't as effective, ultimately, as overt research methods. I tend to agree that covert research is almost never ethical, although there may be times when deception is necessary. But sometimes deception is crucial. Let's say you want to study whether bankers practice racial discrimination in their mortgage lending practices. One of the best ways to do so is to send out matched testers—people with identical credentials but limn different racial/ethnic groups and see if they are treated dillerently when (hey apply for mortgages. The mortgage lender can't give informed i onsent; otherwise, the test won't work. Is this ethical? It probably is ethical to use deception when the intent is to uncover some kind of abuse or bad practice. What about in other situations? Ik-fore choosing a covert research strategy, I would encourage you to think carefully about the dilemmas involved. You might want to consider how you would feel if you found out that a researcher was studying you without your consent. When do you think you should reasonably expect privacy—which presumably includes the right not to be researched? In your home? Your car? Your place of employment? In public places like shopping malls or restaurants? What about in a classroom? In general, when people can reasonably expect privacy, you must attempt to gain informed consent. In public places (like shopping malls), where people reasonably expect that others might see them and in which informed consent might be impossible to obtain, covert research may be fine. Sometimes, beginning researchers think that they must conduct research covertly because if people know they are being observed they will act differently. While this may be true, there are things you can do to minimize the effect. (Chapter 4 will deal with this issue in detail.) You should also be aware that you miss important details when you conduct research covertly, because you are typically less free to ask clarifying questions or to take notes. In addition, the stress of researching covertly may subtly shift your interactions and make you inhibited around the people you want to study. Still, the issue of covert versus overt strategies is not always clear-cut, and researchers continue to debate the issue, often passionately (see Mitchell 1993; Punch, 1998). PRACTICAL CONCERNS IN PROTECTING PRIVACY Researchers need to be careful about protecting participants' confidentiality while collecting and analyzing the data and when publishing it or otherwise disseminating their results. While collecting your data, there are several things you can do to try to protect your respondents' privacy. First, you can make sure that you don't leave their names or phone numbers or e-mail addresses lying about where others can see them. (This might seem obvious, but for those of us who tend to be messy, it's sound advice.) It's a good idea to keep this kind of information in a locked file cabinet. Some researchers, especially those who are working with members of controversial groups, are carelul not to keep lists ol participants, or they disi aid them as soon as possible. If you share a computer with others or use the Internet to correspond with research participants, you should take special steps to preserve your participants' privacy. A relatively easy way to help protect your subjects' privacy when you are writing notes and transcribing interview tapes is to assign a code number or pseudonym and use that instead of the person's real name. (And, if you do attempt to do "high-risk" work—for example, research on groups that break the law—it's absolutely crucial.) My own students have found this important in their class research projects. One group was conducting a campus study on student cheating. They knew that student participants wouldn't be honest if they thought that professors might learn their identities. If the research was going to yield anything of value, the student researchers had to be scrupulous about protecting their informants' privacy. But I also had to read the student researchers' field notes and interview transcripts in order to evaluate their course work. By omitting names from those documents and by asking me not to observe a focus group session, the students were able to protect their informants. You also need to think about how to discuss your work with others while you are doing it. If you are excited about how your data collection is going, it's often difficult not to talk about it. But you need to consider whether you are giving away important identifying details. For example, a former student of mine and I presented a paper based on some ongoing field research in an urban housing project in Kansas City (Masuda 1998). The conference was in Iowa. We didn't reveal the name of the housing project (there were a number in the city), the management company, or any of the people involved. Yet at the end of the session, someone came up to us and asked if one of the participants in the study was "Jan" (I won't give you her real name, but he had guessed it correctly). It sometimes takes very little information for someone even slightly familiar with the setting to guess another individual's identity. This may be even more crucial in a college or university setting if you are conducting a research project on campus. When you are presenting your research—either as a presentation or a published piece of writing—you need to be especially careful to protect your participants. Most social researchers use pseudonyms for participants and vague descriptions for the actual place in which they do research. Thus, William Foote Whyte (1943) wrote about "Cornerville," which he revealed years later to be the North End of Boston. Judith Stacey (1991) located her respondents in Silicon Valley—a large enough geographic region to protect the confidentiality of her informants. I.M.klNG AltOUT I I MKAI ISSUI S I idual issues may occur at all stages of a research project. In the early ,iages, you should plan to minimize any harmful consequences to your i. search participants. You also need to consider what you can give back to i he people you are studying. As your research unfolds, new ethical challenges may arise as your work goes in new and unexpected directions. At all stages, it is important to talk with others about these ethical challenges, i Hhers may see ethical problems you hadn't considered, or they may offer novel solutions. By talking with others, you may gain greater insight into how to handle ethical dilemmas. OUESTIONS FOR THOUGHT 1. Some researchers have gone to great lengths and some have even broken the law to conduct their research. For example, Jeff Ferrell (1995) was arrested for "graffiti vandalism" in the course of his participant observation with hip-hop graffiti writers. In doing research on sex workers, Wendy Chapkis (1997) spent an afternoon selling sex to women clients (legally) in Amsterdam. Much closer to home, some students have gone to parties where illegal substances have been present. How far would you go in your research? Would you break the law? Do something that some consider unethical? Why or why not? 2. What would you do if you observed someone doing something illegal or unethical in the course of your research? Why? What would you do if a research participant told you about doing something illegal (but you didn't observe it)? Would your reaction be different? Why? 3. Imagine that you have been interviewing students about a whole range of controversial issues, including their religious beliefs, their use of alcohol and drugs, and their sexual behaviors. How can you protect their privacy and confidentiality? What if you want to interview students whom you already know? Is this ethical? What kinds of special precautions should you take? 4. Imagine that you are conducting research on sexual violence. You have been interviewing students and observing at a rape crisis shelter. One of your research participants becomes upset during the interview, visibly shaking and crying. What should you do? Suppose that she names someone you know or have heard of as her attacker. What should you do? Imagine that you have been doing an in-ilc|>lh sludv ol peisonal rein tionships in a small community. You have been ohsei viug and ( (inducting taped interviews. Suppose that in one of your interviews your suhjci I freely talks about a number of people whom you know through your research. What should you do? Suppose that in the course of conducting one of your interviews, you become very attracted to the person you are interviewing. Is it ethical to ask him or her for a date? EXERCISES 1. Find out the procedures for getting IRB approval on your campus. Do students have to get special approval for their research projects? What procedures must they follow? 2. Choose one of the research questions you evaluated in Chapter 2. Develop a plan for protecting research subjects' confidentiality and gaining informed consent, if appropriate. 3. Following the model in Appendix B, develop a form for obtaining informed consent from a potential research subject. SUGGESTIONS FOR FURTHER READING Fine, Michelle; Lois Weis; Susan Weseen; and Loonmun Wong. "For Whom? Qualitative Research, Representations, and Social Responsibilities." Pp. 107-131 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage, 2000. Punch, Maurice. "Politics and Ethics in Qualitative Research." Pp. 156-184 in The Landscape of Qualitative Research, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage, 1998. Richardson, James T. "Experiencing Research on New Religions and Cults: Practical and Ethical Considerations." Pp. 62-71 in Experiencing Fieldwork: An Inside View of Qualitative Research, edited by William B. Shaffir and Robert A. Stebbins. Newbury Park, CA: Sage, 1991. Stacey, Judith. "Can There Be a Feminist Ethnography?" Pp. 88-101 in Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana: University of Illinois Press, 1996. Wolf, Diane L. (ed.). Feminist Dilemmas in Fieldwork. Boulder, CO: Westview Press, 1996. Observation: Participant and Otherwise You observe all the time—on your way from home to school and work, during class and meals, and so on. But how much do you really see? If you're like most people, much of what you observe in your day-to-day life you ignore. Your eyes don't really focus, or you don't think much about what you're seeing. You forget almost as soon as you move on. Here's a quick test I sometimes do with my students. I ask them to close their eyes and tell me the color of the walls and the floor of the classroom. I Isually, most of them don't know. Can you remember the color of the classroom where your methods class meets? Probably not. This shouldn't be a surprise. After all, you don't really need to remember its color, and so you don't. But you probably do remember the topic from your last assignment— or at least could bring it to mind if you thought about it. In class, you're expected to focus on the substance of the course, and not on your physical surroundings. You're likely to remember more about the topic because you have been focusing on it. How much more do you see when you really pay attention? How much more can you recall when you make a mental note to remember something or when you take the time to write things down? When you focus on observing, you see so much more. In qualitative social research, you need to train yourself to look, to observe. This isn't that much different from what you do the rest of the time. It's more a question of degree. 57 Observation—--looking in a ionised vvav i'i ;it tin 1h-.ii > ol qualitative research.This chapter focuses on the ways in whu Ii m liolar. use observation to understand social life. Observational studies are veiy dillerent from sin veys or experiments or even interviews. Instead of asking people questions about their thoughts and behavior or conducting experiments in a lahoni tory, these researchers go to the "natural" settings in which social life takes place and observe what people "really" do in those settings. In observational studies, researchers play a much more prominent role than in other kinds ol research. They usually immerse themselves for relatively long periods ol time in a specific setting, getting to know the people there intimately. Researchers have used observation to study many different groups of people in many different kinds of settings. William Shaffir, for example, used participant observation to study ultra Orthodox Jews in Israel anil Montreal (Shaffir 1991). Alma Gottlieb lived for 15 months among the Beng people of West Africa with her husband, fiction writer Philip Graham. They wrote about their experiences in an award-winning book, Parallel Worlds: An Anthropologist and a Writer Encounter Africa (Gottlieb and Graham 1993). Closer to home, researchers have studied lesbian communities in the Midwest (Krieger 1983) and Northeast (Esterberg 1997), low-income women in the Boston area (Dodson 1998), families in California's Silicon Valley (Stacey 1991), and children in elementary schools in California and Michigan (Thome 1993). Researchers have used observational methods to study taxi drivers, jazz musicians, biker gangs, and police officers, among many others. ETHNOGRAPHY Scholars from different traditions use somewhat different words to describe this kind of work. Some, especially those from anthropology, use the word ethnography. Often, those who do ethnography—called ethnographers—supplement their observations with in-depth interviews (which we'll deal with in Chapter 5) and archival research (which we'll consider in Chapter 6). Historically, sociologists have used the term participant observation, although an increasing number call it ethnography as well. Still others call it field research or field studies. Whatever you call it, this kind of research entails immersing yourself in the social life of a group, observing, and writing about what you see. Many of those who have traditionally done ethnography have argued that the study of culture is—or should be—the central concern of this kind of research (Geertz 1973). There are many debates about what ethnogra- l-liv (and i ulliiie) is; one ant In o|iu simply ask questions over the phone. You will be able to see, hear, and niell what the men themselves do and thereby get a much richer under-••< amling of shelter life. Your understanding will be deepened if you actually .I iend the night and experience some of the same things as the homeless men. And if you hang out and observe at the shelter for an extended period ol time—not just once—your understanding will be further enriched. By •.baring experiences with the men over time, you can gain insight into their lives. Will you actually begin to experience life exactly as the homeless men do? Not as long as you have a home that you could go to. But you will come much closer to their experiences than you would otherwise. Ethnography developed out of an impulse for westerners to understand what they thought of as "primitive" people. Traditionally, ethnographic researchers were encouraged to study groups of people in distant communities or, at least, groups of people who were very different from the researchers themselves. Thus, anthropologists typically studied groups in far-off places, and sociologists studied hoboes, slum dwellers, prostitutes, and the like. Ethnographic researchers were not encouraged to study themselves or people like themselves. While ethnographers still study marginalized groups, they often do so with the explicit aim of giving voice or bearing witness to their experiences. Many ethnographers hope that their writing will dispel stereotypes or help increase understanding of the group they are studying. More recently, some scholars have created what they call autoethnogra-phy, or ethnography of the self. The purpose of this kind of ethnography is to try to understand the researcher's own experiences by using introspection and recalling personal experiences (Ellis 1999). Autoethnographers focus on their own thoughts, feelings, and experiences. By exploring their own lives, autoethnographers like Carolyn Ellis (1999) hope to shed light on the lives and experiences of others. Although ethnographers work in a vaiielv <>l sellings ami see them selves as having somewhat dillcront purposes, at heail lesearehers who An this kind of research all do the following: 1. Immerse themselves in a field setting, usually for an extended period of time 2. Participate in a variety of ways 3. Observe while they are participating 4. Take notes about what they are observing 5. Conduct informal (and sometimes formal) interviews 6. Take more notes (and more notes and more notes and more notes) 7. Analyze their notes 8. Write up their analysis, often in the form of a story or extended narrative PARTICIPATING AND OBSERVING When I discuss participant observation with my students, they often focus more on the observation side and less on the participation side. That is, they think they have to spend all their time watching and taking notes, and not much time joining in. They often don't understand what I mean by participating in the life of the group. One research group in my class, for example, studied the parking problem at our university. They spent many hours watching people drive in and out of the parking lots. They took lots of notes while they were observing, but they didn't really talk to anyone while they were doing their observations. No one knew that they were observing the parking lots, and—because parking lots are a public place—no one seemed to care. In this case, the students took primarily an observational role. That is, they didn't interact much with others in the setting, and no one changed their behavior as a result of their observation. But were they really pure observers? That is, were they not participating in the setting at all? Actually, they were participating, for they were all students at the university, and they all had experiences of their own trying to park their cars on campus. Even though they had little contact with others while they were observing the lots, they still brought their own experiences as students and as drivers to their observations. These experiences clearly shaped how they thought about parking on campus, what they noticed while they were observing, and how they interpreted what they saw. 1......e i<•>,( an hers study group.', lli.il they aie already involved in. lor .mple, I'alrhia and Peter Ailler dei ided to study after-school groups 1 nr..- o| their experiences as parents (Adler and Adler 1994). As parents, •>■ u> ie already involved with school-aged children in a variety of roles; ■I. i>liled the role of researcher to study the after-school programs more !■ main ally. Other researchers enter into field settings specifically to 1-, them. Once in the setting, they figure out what role to play and how i i to participate. For example, William Foote Whyte spent several years ■I ■ • ivmg anil participating in Boston's North End during the late 1930s i '■ liyie l')43). Prior to his study, he had had no contacts with the commu-"ic, lie picked the neighborhood he called "Cornerville" for study because i> usis interested in what he then called "slum districts." (In an appendix to '!>• ilmd edition of his book, Whyte wrote about how he came to study ' . 'i nei ville," what roles he played, and how he was able to gain access.) I low much observing should you do, and how much participating? > lule it is possible for you to participate minimally in a setting and for par-ipanls to be unaware that you're observing, most researchers participate "I. ast in some ways. And while it may be ethical to engage in covert obser-iiion.s in very open and public places, in more private places covert " .e.nch may be unethical. Vmi might think of participation as having multiple dimensions (Atkin- • >n and Hammersley 1998). The researcher might be completely known to i In >se being researched, a little known, or completely unknown. Those being ■ • searched might know a little bit or nothing about the research, or they ni.iy be actively involved in setting the agenda for the research, as they Miuld be in field studies conducted from a critical perspective. While in the li. Id setting, the researcher might engage in a variety of tasks, from strictly "bserving (as the students in my parking group did) to being completely ■ ni'.aged in the scene. How much observing and how much participating \od do depends on your own inclinations and on the research setting. The balance between observation and participation can also change over the • outse of a research project, as you become more and more familiar with i he setting and the people in it. I ME SELF AS INSTRUMENT In observational research, you are the research instrument. What's important is what you observe, what you see and hear. Clearly, you cannot observe everything at one time; you have to decide what to focus your attention on. I lowever, unlike in survey research, you do not have an interview guide or schedule to lull back on. There's no list of things that you luire to locus on And humans are moving targets! They don't simply si am I still and wait loi you to record their behavior or jot down what they just said. By the lime you've noted a behavior, the people in the setting have moved on. Sometimes, students think that if they could just record everything either by hand or with a tape recorder—they would somehow have a com plete and objective record of "what happened." As beginning researchers soon discover, however, they miss much of what people say or do. There's simply no way that they can record everything that is going on in the setting. But even if they could somehow magically record everything, they would still have to figure out which of the things they saw or heard were important. In qualitative research, you have to use your judgment to figure out what to focus on. You have to decide what is important in the setting and what you should focus on. At the same time, your own personal qualities—how well you get along with other people, whether you're outgoing or more reserved, whether you have a sense of humor, and so forth—will dramatically shape how people react to you in the field and, thus, what kind of evidence you collect. Your own qualities shape what you can see in the field setting. As we'll discuss shortly, developing personal relationships is one of the crucial tasks of field-work. In this sense, ethnographic research is very personal research. Who you are and what qualities you bring to your work matter. THINKING ABOUT SETTINGS AND SELECTING A SITE One of the first tasks in an observational study is to pick a research setting. Where are you going to observe? What do you hope to see? You also need to consider the physical place in which you plan to conduct your research. As you are considering potential settings, you need to ask these questions: ♦ Is this an appropriate place to study what you want to study? * How can you define the boundaries of your field setting? # Who are you in relation to this site? What kinds of connections do you already have to it? •» How can you gain access to this site? ♦ What kinds of ethical dilemmas will you need to consider in using this site? What kinds of risk are inherent in this setting? ■ in Is I liis an Appropriate I'lacr lo Study Wlial You VVanl lo Study? < >ne of the Iirst things to think about is the kind of setting that will be ..ppiopriate for your topic. Some kinds of topics are more amenable to ..I.\ei vational study than others. For example, if you want to know how the ii.. .lia depicts urban teenagers, you probably shouldn't do an observational in. ly. You'd be much better off doing a systematic study of the media, such .. newspapers and TV programs. But if you want to know how teens them-..■Ives think about their lives, then you might do participant observation th a group of urban teenagers—at school, in their after-school hangouts, their jobs, and at home. Suppose you want to know what women who receive public assistance iliink about their experiences. It wouldn't make sense to observe the social workers who administer benefits. While that may end up being a part of y< ii ir study, you would be much better off beginning with the women them-.. Ives. But if you are interested in how social workers think about their . lien Is, you will want to find a location where social workers interact. In general, good sites for participant observation tend to be relatively all-scale. You couldn't do a good job observing an entire country, for .ample, or even a state. But you could observe a small town, a relatively .ell-contained neighborhood within a city, or a workplace. At the same time, good settings should also include some kind of face-to-face interaction. Although it is increasingly possible to observe virtual communities on the Internet, these studies can be problematic (Rutter and Smith 1999). Settings without any people in them do not make good sites for observing social behavior!* Ml .'\a What Are the Boundaries of This Site? ()nce you've established a general setting, you need to consider the boundaries of that setting. Where, exactly, do you intend to observe and participate? Will all your observations be made in one single location, or will you participate in a number of places? Some topics seem to suggest settings with relatively clear boundaries. For example, one group of students I worked with was interested in the question of how (and what) people ate in public. They wanted to see if people ate differently when they were with *Of course, a great deal of information about social behavior can be gleaned from physical l races or archival materials even when living people aren't present, But these kinds of materials are studied using somewhat different strategies. people than when they were alone, ami they wanted l<> see how different groups of people negotiated dining halls. Their held setting was relatively straightforward: They observed in the school cafeterias, a relatively well-bounded place. Other settings seem more amorphous. For example, if you are interested in how social workers think about their clients, you might want to spend some time observing in social workers' offices. Yet much of the work that social workers do is carried out in the field. Therefore, you might also ride along with the workers as they go about their daily business. In addition, you might attend any gatherings outside of work, such as informal social get-togethers or special events, that the social workers engage in. In this case, your field setting may include many different physical sites. As you begin your fieldwork, you should try to define as clearly as you can where you will be observing and what you will be doing. Although the boundaries of your setting may very well change, as will your level of participation, it is best to begin with at least a general idea of what you hope to be doing. Who Are You in Relation to This Site? You also need to consider what personal connections you have to the setting. What do you already know about this place? What kinds of stereotypes might you have about the place and the people in it? Do you already know people in the setting? Do they know you? What do you think they think about you? Does the setting have any personal meanings for you? Does it evoke any particular emotions or feelings? Some people suggest that you shouldn't study a place you are too familiar with. They argue that your familiarity with the setting will make it hard for you to see anything new. This can certainly be a danger. For example, one of the students in my parking lot group had difficulty seeing that there was anything to study in her setting. She didn't really believe that there was anything new to see, and so her first field notes were very thin. A bunch of cars were parked, she noted; some came in, and some went out. She wrote that she didn't expect to see anything, and indeed she didn't. Being familiar with a setting can give you some hints about where to begin. For example, if you already volunteer at a homeless shelter, you may have an idea of what's important to the people who stay in it. The shelter residents may also feel more comfortable about your presence in the shelter and your research role. You may have access that you wouldn't be able to get otherwise. However, the shelter residents may have some preconceived notions about your role. They may have a hard time thinking of you as a le'.eau her, and not as a volunteer. II yui det. ide to do your rest-arch in a set-hug where others already know you, you need to think about how they may see you, for that will certainly affect your research. lanel Theophano and Karen Curtis wrote about their experiences .lomg heldwork in an Italian American community (1996). In studying the i. I.oionship between food and ethnicity, they shared family meals in their held site over an extended period of time. One of them, Theophano, was 11 tarried and had a child; the other, Curtis, was single. They found that they were treated differently in the field because of their different statuses. Ilei a use she was a mother, Theophano was "expected to eat the leftover looils that mothers eat; in contrast, [Curtis] was seen as a daughter and provided with newly prepared foods at each meal" (Theophano and Curtis 1996, p. 159). You should also consider whether the setting has any personal meanings that may influence your study. For example, if you have been homeless yourself, you may find that studying a homeless shelter evokes powerful emotions. The existence of these feelings isn't necessarily a signal that you shouldn't do your research in the setting. Sometimes, these emotional cues can help you uncover profound truths about your setting—and yourself. But vou should at least consider whether you want to put yourself in a position to have those experiences. If you do decide to proceed, you should be careful to distinguish your own feelings about the setting from those that the participants themselves may have. Although they may be similar to yours, they may also not be. i low Can You Gain Access to This Site? I'.specially for students, who may only have a semester or two to complete a project, the question of access is paramount. You need to think about whether you can gain access to the setting. Whom might you need to ask for formal permission? Do you have connections that can help you gain access? Who are the gatekeepers, the people who can make your entry easier or stop your research completely? In public places where you don't intend to announce that you are observing, such as shopping malls or parking lots, you may be able to observe without asking formal permission. Access in these cases is easy: You simply enter like any other member of the public. Thus, the students in my parking lot group merely sat in their cars or at the edges of the lot and observed. Similarly, in her research on public spaces such as airport waiting areas, Lyn Lofland simply sat down in the waiting rooms like any other passenger (Lofland 1985). Although it is possible" to do covert research In nioir piivale spaces, you need to consider whether it is ethical to do so. (('liapter J discusses some of the ethical concerns.) You should also be aware of the stress that doing covert research can place on the researcher—and the likelihood that a covert researcher can actually "pass" in the setting. Richard Mitchell discussed his experience as a covert researcher among right-wing survivalists (1991). He recounts: "We had come as covert researchers, hoping to blend in. We arrived driving a late-model diesel-poweretl Peugeot station wagon, 'disguised' for the occasion in freshly pressed discount-store duck-hunting outfits over preppy L.L. Bean pants, Patagonia jackets, and Nike trainers." Rather than blend in, he found that "our disguises were taken as signs of naive enthusiasm. (Who else would wear such ludicrous costumes?) We were accepted, treated with gentle respect, even praised. But we were never overlooked" (Mitchell 1991, pp. 105-106). Yet clothes were not the least of it. One evening, Mitchell found himself in a small group doing guard duty at a Christian Patriots Survival Conference. As the group began swapping violent racist and homophobic stories, Mitchell found himself in the difficult position of having to tell one himself in order to be accepted. He closed his account of the event in this way: "If there are researchers who can participate in such business without feeling, I am not one of them nor do I ever hope to be. What I do hope is someday to forget, forget those unmistakable sounds, my own voice, my own words, telling that Nine O'clock story" (Mitchell 1991, p. 107). In most cases, you need to ask formal permission before you can do research in more private places. And even in some public places, if you are going to observe over an extended period of time, you would do well to ask permission. (Your professor or the institutional review board at your school can help you determine what is appropriate for your own research.) Before entering the field, you need to determine who can grant you permission to study a particular site. You may find that you need to go up several levels in the hierarchy. Suppose you want to observe a classroom in a school. Perhaps you know the teacher and feel you have a reasonable chance of getting her permission. She may refer you to the principal, who might, in turn, refer you to the superintendent of schools. The superintendent may decide that you need to ask permission from the parents before you can begin. At each level, it may take weeks or even months to gain approval to do the research—and you may ultimately be turned down. Sometimes, the person who can formally grant you permission may not be the same person who can actually gain you access to people within the setting. Often, people act as gatekeepers. Because of either their formal position or their personal characteristics, they can influence others in the set- iiug lo either accept you or not. In many ol'lices, lor example, secretaries and . .thei clerical workers are the ones who can ensure- that you actually get in io see executives. In an informal setting, such as a group of friends, one per-,on may be central, an informal leader. In an unfamiliar setting, it may take voii a while to figure out who the informal gatekeepers are. You may also find that choices you make early in your fieldwork affect i he kinds of relationships you can develop later on. For example, imagine i hat you are studying worker relations in a factory. If you are introduced to workers by a boss or manager, they may not trust you and think you are i here to spy on them. If you are introduced by a union steward, they may dunk you are a part of the union, and those who are distrustful of the union may not trust you. Although it is possible to overcome these early impressions over a long period of observation, you should carefully consider how early choices may shape later access. You also need to think about your own personal characteristics and their potential effect on your access to the site. An African American or Asian American researcher, for example, will have enormous difficulty gaining ac-< ess to racist groups like the Christian Identity movement or the Ku Klux Klan. Women might find that they have very different kinds of access than men to college football players or fraternities. Similarly, men may not have the same kind of access to feminist action groups, sororities, or women's athletic teams as women do. Does this mean that you can never study groups of people who are different from you? Certainly not. But it does mean that you need to think about how your personal characteristics might shape how those in a field setting view you and what kinds of access you might (or might not) be able to get. What Kinds of Ethical Dilemmas Might Arise in This Site? When evaluating your field setting, you need to consider whether there are any special ethical issues. We've already talked about the ethical considerations involved in choosing to do covert research, but there may be other issues as well. If there is any possibility that participants could be harmed by your work, you need to make sure that you minimize it. For example, if the researcher who studied radical environmentalists (discussed in Chapter 3) had released the names of his research participants to the federal grand jury, they might have faced criminal charges (Scarce 1995). Sometimes, simply bringing attention to a group might change how it functions, perhaps in negative ways. If the harm to participants is great, you may need to abandon the research project. Some Held researchers have chosen to do theii icseaii h anion); people who are breaking the law, like Daniel Wolf who rode with "outlaw" bikers (Wolf 1991), and Patricia and Peter Adler, who studied drug traffickers (Adler 1985; Adler and Adler 1991). In such cases, you need to think about your own limits in terms of the types of activities you feel comfortable observing. If you observe an illegal activity as part of your research, what should you do? Are you obligated to tell someone? Under what circumstances? While you might think that this kind of ethical dilemma is farfetched or confined only to professional researchers, student researchers encounter it, too. For example, my students are often interested in studying student drinking on campus. What should they do if they encounter, in the course of their research, underage students drinking? Should they writedown what they saw in their field notes? What if they see someone drinking and driving or someone who seems to have a serious problem with alcohol? Although you cannot predict in advance all of the ethical problems you may encounter, you do need to think through the kinds of issues that may arise in a given setting. What Kinds of Risks Are Inherent in This Setting? Finally, you need to evaluate the risks inherent in the setting you have chosen. In most settings, the potential risks are relatively minimal. But in some field settings, the risks are substantial. Again, Adler and Adler's study of drug traffickers and Wolf's study of motorcycle gangs come to mind. Before you enter into a setting, think carefully about whether there are any personal risks and whether you are willing to take them. GETTING IN AND GAINING ACCESS Once you've settled on a field site and obtained formal permission to begin your research, what do you do next? Every field site is unique, and so a textbook like this one can only give you general advice. You might find it useful to read stories that other researchers have written about their early days in the field. (See, for example, the accounts published in Journeys Through Ethnography or Experiencing Fieldwork; the full citations are given at the end of this chapter.) "Hanging out" is typical behavior of ethnographers. But in any given setting, what does hanging out entail? The early days in an observational research project are often puzzling and can be overwhelming—even if you are studying a site with which you .in- lauuliai. It's not necessarily 'leai what you should be Incusing on or what you should be doing. I wen if you have already selected a general lopi,, it can be difficult to figure out exactly what you should be observing, loi example, let's say you're interested in how kindergarten-age children develop social relationships. You've received permission to observe in a . lassmom setting. When you first go to the classroom to observe, where do you situate yourself? What vantage point should you take? Do you sit still or move around? If there is a lot of activity going on in the classroom (as is typical in such settings), where should you direct your attention first? What kind of role should you play? How do you explain what you're doing to the i hilarem Developing Relationships (iaining trust and developing relationships with the people in your setting is your main task. Especially in the early days, people may not trust you or pay much attention to you. And if they do notice you, they may wonder what you are doing there and not let you "in" to observe at all. Hanging out, after all, is rather unusual—and even suspicious—behavior in many settings. Again, a textbook can give you only a very general idea about how to develop relationships. Certainly, in general, it's better to be courteous and polite than rude. It's better to act interested than bored. It's better to listen to others than to talk about yourself. It's better to be unobtrusive than splashy. But in specific field settings, you still need to figure out what role you should play. Some ethnographers play the "naive incompetent" role. That is, they adopt the persona of someone who is nice but doesn't really know very much about the people in the setting and is willing to ask what might seem like dumb questions. The goal here is for people to take pity on the researcher as someone who needs instruction. In any human relationship (research-related or not), there is usually give-and-take. As a researcher, you are taking up people's time. You are asking them to give you information and to help you out in your research. Why should anyone take the time to talk with you or let you follow along? Sometimes, people appreciate the opportunity to talk about their lives; they feel that they have an important story to tell and that the researcher can help them do that. Other times, ethnographers can do small favors: giving rides, buying meals, and so forth. Patricia and Peter Adler talk specifically about reciprocity and exchange (1991, pp. 175-176). They note that researchers "can turn the norms of reciprocity to their advantage, offering assistance to subjects (or potential subjects), thereby hoping to build leeltngs ol'trust and indebtedness" (Adler and Adler 1991, p. 17(i). In theii study ol'drug traffickers, they offered small loans, babysitting, and dinners. Whatever role you play, it's helpful to remember that the people in the setting are the true experts on their own lives. As an ethnographer, you arc trying to learn from them. You can learn more by showing an interest and by being open to what they have to say and to show you. Cultivating Informants As you spend time in your field setting, you may find yourself developing special relationships with certain individuals. Some individuals may seem more knowledgeable about the scene and be more willing to share their insights with you. Others may take you "under their wing," so to speak, and help you gain access. These individuals are sometimes called informants (or key informants, for those who play a particularly special role). In his study of Boston's North End, William F. Whyte (1993) described the relationship he had with a key informant named Doc. Until he met Doc, he was unable to get any insight into "Cornerville." Once Doc introduced him around as his "friend" and was able to vouch that he wasn't a government official, he was able to enter places that had formerly been closed to him. It is important to remember that each informant will have her or his own perspective on the scene and access to somewhat different groups of people. They may, of course, even lie to a researcher. Thus, it's usually a good idea to try to develop relationships with more than one key informant. Developing a Cover Story It's important to create a relatively brief description of your research so that you can tell people what it is you're doing and why you're hanging out. Most people will simply be curious about why you're there and what you're doing, but they don't usually want an extended response. (Think about it: When you ask someone, "How are you?" do you really want a long, detailed response? Probably not.) It's important that your description be appropriate for your audience. For example, a teacher may want a more extended version than kindergarten students will. Should you tell people exactly what it is you're studying? It depends. Usually, people in the field won't want to know all the details of your research, although sometimes they may. There are also instances in which telling people exactly what you're studying might negatively affect the research. For example, let's say you're interested in gender relations in the i lassroom spei ilically, in whether the teacher treats male and female students differently. II you tell the teacher exactly what you're interested in, .he may change her behaviors. In that case, you may want to provide a more general description of your work ("I'm interested in student-teacher i elalionships"). Should you ever tell an outright lie about the purpose of your research? Probably not. In doing fieldwork, your aim is to develop relationships. If you have lied to the people in your setting, how can you expect them to tell you the truth? Nevertheless, researchers sometimes find themselves telling "polite" lies or lying by omission—for example, not telling details about the purposes of the research or not disclosing details about their personal lives. This can be a source of tension. As a researcher, how much should you disclose about yourself? After all, your research participants might disapprove o I some aspect of your identity or social life that you think is not relevant to the research. Sometimes, researchers know that they will be critical of the group in their write-up. This, too, can be a source of tension. In fact, it is just this kind of tension that led William Shaffir and Robert Stebbins to describe lieldwork as one of the "more disagreeable activities that humanity has fashioned for itself" (1991, p. 1). Becoming Invisible Won't people change their behavior if they know that a researcher is watching them? Most of us don't like outsiders to see us doing something socially unacceptable. If we knew that someone were watching us, we would probably behave much better than we would otherwise. In a classroom, for example, a teacher might pay special attention to her lesson plans if she knows she is being observed. The children might act differently as well—perhaps more or less well behaved. Because people will change their behavior, my students often argue that covert research is better. If people don't know you're observing them, they won't change. But this isn't necessarily the case. In ethnographic research, the researcher typically stays in the field for an extended period of time. Over time, those in the setting become habituated to her or his presence. People often forget that the researcher is actually gathering data while hanging out. The researcher may be of so little importance to those in the field that they simply don't attend to his or her presence. Or the researcher poses no threat to their everyday activities, so they pay her or him no mind. While those in the setting may never accept the researcher as one of them, they can come to accept the role that she or he plays in that setting. Besides, it's difficult for people to be on their best behavior for very long periods of time. II'you spend enough time in the held, it's likely that people will eventually come to behave as they ordinarily would, or at hast approximately so. Dealing with Emotions Emotions are part of all life experiences, including field research. As you go about your fieldwork, you'll likely experience various emotions, including boredom, disgust, anger, and pleasure. You may feel anxious about being accepted by your informants or about figuring out your role in the field. You may become bored, as my students who observed the school cafeterias eventually did. You may become angry at or feel manipulated by your informants, as Patricia and Peter Adler sometimes did in their study of drug traffickers (1991, p. 170). Researchers in the positivist tradition typically have argued that emotions are too subjective and have little place in social research because they affect researchers' objectivity. But other researchers argue the opposite: that it is crucial to pay attention to your emotions when you are conducting qualitative research. Sherryl Kleinman (1991, pp. 184-185) suggests that if you don't pay attention to your emotions they will still shape the research— you just won't know exactly how. Emotions provide important clues to what is actually happening in the field. For example, in doing field research with a support group of at-home mothers, I consistently felt anxious or guilty about my own parenting in the presence of one of my informants. A careful analysis of my field notes led me to an important insight into insider-outsider dynamics within the group. WRITING FIELD NOTES Developing Observational Skills How do you learn to observe? How do you learn to remember what you have seen? And how do you learn to make sense of what you are observing? Like all skills, observation can be learned. You get better at it by practicing. As a first step, I recommend that you work through some of the exercises at the end of this chapter. You might also be interested in a 1998 book by Valerie Janesick, "Stretching" Exercises for Qualitative Researchers, that provides a number of activities to help beginning researchers become better observers (see the "Suggestions for Further Reading"). Finally, the practice of writing field notes can help you develop your observational skills. In fail, one of the most iuipoilant tasks you will engage in during field i,•search is writing held notes. Why write held notes? For one thing, writing things down helps you to remember them. Because field research usually lakes place over many months, if not years, you simply cannot remember everything you see without writing it down. Many field researchers argue that if you don't write something down it's almost as if you didn't observe it. The field notes become your written record of what you observed. In addition, writing is a way of making meaning. You begin to organize your thoughts and make sense of what you are observing by writing about it. Thus, as Robert Emerson, Rachel Fretz, and Linda Shaw argue, writing held notes "is not so much a matter of passively copying down 'facts' about 'what happened.' Rather, such writing involves active processes of interpretation and sense-making: noting and writing down some things as 'significant,' noting but ignoring others as 'not significant,' and even missing other possibly significant things altogether" (1995, p. 8). As you write, you make judgments about what is important in your setting. When to Write? Should you write in the field or should you wait until you are out of people's presence? Some researchers find it awkward to take notes in front of those whom they are researching; they feel that taking notes in the field interrupts the flow of action. Sometimes, note taking is clearly inappropriate. For example, if you are observing a very emotionally charged event (such as a funeral or a wedding), it might seem rude or disrespectful to take notes. At other times, and in other settings, it seems perfectly natural. For example, if you are observing in a school cafeteria or a shopping mall, it may feel perfectly appropriate to take notes. Some researchers who take notes openly in the field argue that it gives them more freedom to ask questions. It may also reinforce, in the minds of participants, their role as researcher. Whether you take notes openly or not, I recommend that you carry a small notebook with you and try to jot down key words or phrases while in the field. If you're in a setting in which you feel it is inappropriate to take notes, you can sometimes leave the scene briefly or take a quick break to jot key words. What should you jot down? You can use mnemonics (words that will help you to remember), memorable quotes, and a few details about the interaction. Some researchers carry a tape recorder with them. Then, on their drive back from the field setting, they record as much as they can remember. Immediately upon leaving the field, whether you have recorded notes afterward or jotted down notes in the field, I recommend that you write up lull held notes. These arc your lullesl, most complete ilesi nplions ol whal occurred during your period ol observation, including (lie events, the se( ting, and the participants.They may also include your initial sense ol what is happening in the setting or what the events mean to the participants, liven if you use a tape recorder or are able to write fairly complete notes in the field, I still recommend that you write up full field notes as soon as possible after leaving the field. What to Write? Your notes should reflect your growing understanding of the social scene you are studying. You should take notes about the physical setting and the people within it, as well as the behaviors you observe. When you first enter a field setting, you should take notes about your initial impressions. You might want to draw a map of the site as well. Then jot down notes about what the place looks like, what kinds of things immediately stand out, what the people look like, how they are dressed, and how they interact with each other. Although your impressions are likely to change as you gain experience in the field, these initial impressions are crucial. After you have been in the setting for a while, you will become habituated. That is, you won't see certain things that you take for granted, just as you don't really see the color of the walls and floor of your classroom. The following is an early field note on the "Parker Gardens" Section 8 Housing Complex, on which Kazuyo Masuda based her master's thesis (1998). (This study was described in Chapter 2.) Notice how this excerpt helps set the scene. It describes the housing complex in which much of the action takes place but does not provide details about any action.* Parker Gardens is a complex of two-storied apartments, covering two by two square blocks, or four blocks in total. There are no through streets cutting through the complex, but parking lots in the middle and on the west side. The complex starts at V St. on the east, and climbs the street, up 23rd and 24th, to the next north-south street, the one with Jefferson High School on it. The main floor of the buildings is brick, and the second floor is wood siding, with the kind of early 1960s window trim that is a three-sided curb around the win- *A number of researchers were involved with this research in the early stages of the project. Anthropologist Ken Erickson, Kazuyo Masuda, and I all contributed field notes. This excerpt was contributed by Ken Erickson. dovvs. The buildings aie in m-ed ol paint, some have screen doors and some do not The management ol In e and the "community room" next In (hem are in one ol the buildings, apparently with an occupied apai (inenl above them. (We could hear footsteps this morning, Ms. R. and I, and she mentioned the four-bedroom apartment above us.) sometimes, you'll want to describe people as well, as Masuda did here: Mr. M. wears expensive socks and expensive loafers and expensive silk shirts and he drives a Mercedes. He wears a gold bracelet and a rather big gold ring, too, and he has round, wire-rim glasses. In your ongoing field notes, you should note what people say and do and what the things they say and do seem to mean to them. When possible, you should provide direct quotations or paraphrases of what people say. Sometimes, you'll write scenes or stories. (For an excellent discussion of the kinds of things you might include in your field notes, see Writing Ethnographic Fieldnotes.) The hardest thing for most beginning researchers is providing enough detail. You probably cannot have enough description in your held notes. The following excerpt, also from the "Parker Gardens" project, tells the story of a tutoring session for children from the housing complex that never really got off the ground. Masuda wrote the following notes after her first day as a volunteer tutor. It is a Monday, shortly after school has begun in the fall. Most of the action occurs between herself and "T," the service coordinator for the building. As I was walking up to the office, I saw children running around. When I walked in the office, Mr. M, T, and another man were there. They were discussing the adult education program. T. told me since this was the first day of the after school program, no kids would show. . . . When we walked into the meeting room together, a man and a woman with a child were working with another man. They appeared to be in an adult education program. There were two loaves of bread, peanut butter, jelly, juice, muffins, cups and napkins laid out on the table. However, T. began to put them back in the refrigerator as soon as we were there. He said, "It's the first day, you know." He and I chatted about school. . . . After the conversation, he said to me, "You can help them [the kids at the complex]." I offered to stay until 5 p.m. He said, "Well you don't need to. We will see what happens between now and 4 p.m." T It'll lor somewhere. I snl down and began in i, ,u\ my hook. A middlc-school-agc hoy tame in ami tiptrittl 111 • ieh igeiatot and looked in. He left, then another hoy tame in and did the same. Alter this one left, a school-age girl came in and did the .same. T. came hack and the two boys came with him. Two girls (one elementary school age) wandered in. "Get the cups out," said T. The kids got the juice out and drank. One boy said, "This is good!" T. began asking the kids questions about their homework. First girl said, "We get homework twice a week in creative writing class. Other classes, it depends." Another girl mumbled something I could not hear. T. told the children to bring their homework next time. "And if you don't have homework, bring something else." Then he realized the kids were drinking up all the juice and told them, "It will be there tomorrow." He put the cups back in the cabinet. The boys were talking all the meanwhile. T. asked them, "Then Tuesdays and Wednesdays are not good days [for the after-school program]? Did you go to school today?" First boy answered, "Yeah.'The kids began discussing which school they go to. Second boy said he was kicked out from one school and did not get to go to the school he wanted to. . .. T. said he would have a program for them on Tuesdays and Wednesdays. Second girl argued: "Tuesdays and Thursdays, that's what on the flyer." T. said, "NO! I wrote it. I know.. .." She ran to the office bulletin board and said, "Look at this."T. and the kids went over. I heard paper being ripped from the bulletin board and tossed in the trash can. After the kids left, T. came back and said to me, "I will have something lined up for you next week. You can go if you want to." He left. I thought about staying longer. I decided not to stay and walked out shortly after. I saw T. driving off as I was leaving the office. There were several children playing around the apartment building. Arguably, not much is happening here. But after observing in the field for an extended period of time, Masuda noticed that many of the programs and initiatives begun by the service coordinator ended in this way: in little actually happening and few children becoming involved. The observation that children were playing outside as Masuda both entered and left the scene, while a small detail, is important. It's not for a lack of children that the after-school program falters. What should you not write down in field notes? Even though it's often difficult to avoid, you should try to avoid generalizations (Emerson, Fretz, and Shaw 1995). For example, let's say you are observing people eating in I a . aleieiia. You shouldn't say that a icseanh pail it ipant "always" gets a ii.iv You tan only report what you observed today: that in this instance, ...m,one (let's call him "X") took a tray. You should also try to avoid judgmental .statements, such as "X eats a lot" or "X is a glutton." Rather than miei■preting his or her behavior, write about what X is doing: "X took four sandwiches, a bowl of soup, and a large bag of chips; he then returned for i wo cups of milk, an ice cream, and a package of cookies." Finally, try to keep your opinions out of your jotted notes. "X ate very quickly, using both hands. Crumbs dropped on the floor and down his shirt" is better than "X ale like a pig." Avoiding generalization is difficult—even for experienced researchers. I or example, go back to the field note describing Mr. M. on page 75. What . l< h's "expensive" mean? To one person, an expensive pair of shoes might be bought at Walmart.To another, expensive shoes might mean a designer label pair bought in a department store like Bloomingdale's or Macy's. I low to Write? You should date each field note and note the time and location in which the observation took place. It's often helpful, too, to list the participants. (Of course, sometimes there are too many to list or the participants change; you will need to decide what is helpful to you.) You may want to use pseudonyms, initials, or code names for your study participants. I always recommend writing full field notes using a word processor and placing them in a three-ring binder. This may help you in your subsequent analysis. Use quotation marks for direct quotes—things you heard people say and were able to write down exactly (or almost exactly) as they said them. Use paraphrases when you can't remember exactly how someone said something but do recall the gist of it. In addition to recording what happened, I also encourage my students to write about their own feelings and initial hunches about the data. But I recommend that you separate your own feelings and impressions from more direct observations. When I write field notes, I usually use brackets to separate my own feelings and thoughts from my actual observations. Some people like to write in two columns—one column for their thoughts and feelings, the other for their observations. Others prefer to use different typefaces or colored type to distinguish the two. People use a variety of styles for their field notes. I generally try to use complete sentences, and I usually start chronologically (what happened first) but then shift around as my writing sparks new memories. Sometimes, writers will begin with a big event anil then move on to smaller, less impoi tant ones. You'll develop your own style as you write. What's most impoi tant is that you keep on writing. lion is possible than in more distant kinds of research, such as briel surveys. In exiling the lield, you need to think carefully about the kinds of expectations your informants may have of you—and you of them. LEAVING THE FIELD At some point, you need to stop observing and leave the field. When should you leave? Ideally, you should leave when you feel you have little more In gain by observing longer. Perhaps your field notes seem to be saying the same things over and over, or additional informants seem to be telling you the same kind of things. Perhaps you feel stale or bored in the setting. Or perhaps you feel that you have learned all you can, at least for the time being. If this is the case, it's probably a good time to leave. In most cases, however, external factors will play a major role in determining when yon leave the field.The summer ends, and you need to go back to the classroom. Or the semester is over, and the final paper is due. Or the funding has run out. Ideally, you should stay in the setting until you feel it's time to go. Mori' practically, you will find that you often leave prematurely. As you prepare to leave the field, you need to consider your future relationships with those whom you have been researching. Will you want to return to the setting later for additional study or clarification? If so, you should be sure to leave on good terms. A general rule is that ethnographers shouldn't "spoil" the field—make it impossible for future researchers to return. Some people believe that researchers should share their findings with their informants or solicit their comments on the final analysis and write-up. Judith Stacey (1991) allowed her informants to have—literally—the last word. As the last chapter in her ethnography of postmodern family life, she printed an interview with one of her informants on her reactions to the text. You also need to consider the obligations you may have incurred during the course of your study. Have your informants come to rely on you in any key ways? If so, can you continue to provide the kind of support they have been getting from you? Or can you find alternative ways for them to get it? The relationships that ethnographers develop with their informants often are in many respects like friendships. Some researchers see themselves as friends with those they are studying (and vice versa). But there's a peculiar imbalance of power in these relationships, for the researcher can leave whenever she or he wants to. The informant, in contrast, typically cannot leave her or his own life. Judith Stacey (1996) argues that, because of the greater intimacy possible in ethnographic research, much greater exploita- i OCUSING YOUR OBSERVATIONS I low do you begin to figure out what is important in your setting? How do vmi begin to narrow your observations and make sense of the relationships you are documenting in your field notes? As in any other kind of research, your data will not "speak for themselves." You have to interpret—figure out—what the evidence means. Researchers often leave this as a last step. They focus for an extended period on gathering data and observing in the held. Then, on returning "home" from their fieldwork, they begin to analyze their materials and develop a final, written report of their research. Because ethnographers typically generate hundreds, or even thousands, of pages of written data (field notes, interview transcripts, documents, and so forth), analysis can seem a daunting task. It's best not to wait until you have finished your observations before beginning to make sense of them. If you begin your analyses while you are still in the field, you can focus your observations on the questions that arise during your analysis. Chapters 8 and 9 deal specifically with data analysis. 1 or now, however, I'll discuss some ways that you can begin to make sense ol the materials you have collected. Making Meaning I like to think of data analysis as answering the question "So what?" Why is your research interesting or important? Why should people care about it? What is the larger sociological significance of your study? Even something as mundane as parking cars can have sociological significance. My parking lot research group, for example, came to argue that at least part of the parking conflict on campus was about social status. Students complained about not being able to find convenient parking spaces. As customers—people who were paying for their education—they felt that they deserved access to good parking spots as part of their tuition and fees. They argued that their tuition, in effect, pays faculty wages; thus, they felt entitled to better parking spaces than the faculty. Faculty members, in contrast, felt that they deserved better parking spots because they had higher status than students. This kind of analysis is called interpretation. Not only do you have to write good descriptions in field research, you also have to interpret what the evidence moans. I low do you begin? I recommend dial, allei you liaye been observing for a while, you read and reread your held note-, ami write memo', or notes to yourself (We'll discuss memos in greater detail in Chapter H ) Try to read with a questioning mind. Then, jot down your though Is, hunches, and feelings about the ongoing research. Sometimes, these memos will help you to make sense of what you're observing. Other times, they will help you explore how you're feeling about your research. Sometimes, they may point you in a new direction. Don't worry if at early stages you seem to be bouncing around among many different ways to think about your material. What these early explorations do is help you tease out different possibilities. Coming to See Multiple Realities At heart, observational research is about looking—and seeing—from multiple perspectives. No two individuals have exactly the same understanding of social reality. Anthropologist Harry Wolcott gives a good example (199!), p. 66). He invites us to consider how a real estate developer, a hunter, and a biologist might appraise a prime piece of rural property. The developer might focus on the possibilities for construction and development, the hunter would mainly be interested in the possibilities for hunting and recreation, and the biologist would be more interested in the flora and fauna. In all three cases, the piece of land remains the same. But how people think about it—what they see in it—differs. As social observers, your task is to try to understand the multiple ways in which individuals perceive reality. QUESTIONS FOR THOUGHT 1. Imagine that someone (maybe a teacher?) has been covertly studying you. This person has been taking notes about how you dress, what you say, how you interact with others. The notes are fairly detailed and sometimes embarrassing, because you didn't know you were being observed. How would you feel about this? Would you feel any differently if you found out that the research was going to be published? Would you want to be able to have a say in what was said about you? 2. Some people argue that ethnography should be more objective, like the natural sciences. Others think it should be more subjective, like the humanities. What do you think? Why? i \l iu isis I < unsider the following Held sellings. What strategies might you use to gain access to the site? What difficulties might you encounter? a. Recruits in the police academy b. A community of recent immigrants i. A disability rights group d. A student organization e. A women's soccer team ' Now imagine that you are a member of another racial/ethnic group. (If you are White, imagine that you are Asian American or African American, for example. If you are Asian, imagine that you are African American or Latino/a, for example.) How might your strategies for accessing the groups in Exercise 1 change? Would access be harder or easier? In what ways? ( Now imagine that you are a member of the other gender. Again, how might your strategies for accessing the groups in Exercise 1 change? Would access be harder or easier? In what ways? ■I. Practice describing yourself to (a) a teacher, (b) a parent, and (c) a friend. How do your self-descriptions change for each audience? What do these self-descriptions say about how you "really" are? V Have a teacher or friend place a few inanimate objects in the middle of a table or desk. Sitting in one place, spend 5 minutes observing and writing down what you see. After the 5 minutes is up, move to another location, and again spend 5 minutes observing and writing down what you see.* What did you see from each perspective? <). Go to a public place where you can observe without disturbing the scene (for example, a waiting room, cafeteria, or mall). Spend about 15 minutes observing and taking notes. Then go home and write up full field notes. Try observing again in the same place at least one or two additional times, each time writing up full field notes. How do your observations change over time? 7. Find a setting where there is a lot of social interaction (for example, a party, a ball game, or a public park). Try to observe intensively without taking notes. Remain in the setting for at least 30 minutes. When you return home, write up full field notes. What are you able to remember? *This exercise is adapted from one in Janesick 1998. SUGGESTIONS I OK I UHINIIl HEADING Burawoy, Michael, et al. Ethnography Unbound. Berkeley: University of California Press, 1991. Emerson, Robert; Rachel Fretz; and Linda Shaw. Writing Ethnographic Fieldiwirs. Chicago: University of Chicago Press, 1995. Janesick, Valerie. "Stretching" Exercises for Qualitative Researchers. Thousand Oaks CA: Sage, 1998. Lareau, Annette, and Jeffrey Shultz (eds.J. Journeys Through Ethnography: Realistic Accounts ofFieldwork. Boulder, CO: Westview Press, 1996. Lofland, John, and Lyn Lofland. Analyzing Social Settings, 3rd ed. Belmont, CA: Wadsworth, 1995. Shaffir, William, and Robert Stebbins. Experiencing Fieldwork: An Inside View of Qualitative Research. Newbury Park, CA: Sage, 1991. Interviews Interviewing is rather like marriage: everybody knows what it is, an awful lot of people do it, and yet behind each closed front door there is 11 world of secrets. —Ann Oakley (1981, p. 30) H. Lave you ever been interviewed? If you have a phone in your name or have spent much time in a shopping mall, chances are, somebody has tried to interview you to get your opinion about some product or service. Maybe you were interviewed as part of the process of getting into college or obtaining a job, or maybe you have been interviewed by a local newspaper or television program about an accomplishment. Valerie Janesick defines an interview as "a meeting of two persons to exchange information and ideas through questions and responses, resulting in communication and joint construction of meaning about a particular topic" (1998, p. 30). There are many different kinds of interviews, each with somewhat different techniques and purposes. Journalists conduct interviews to get information for a news story. Market researchers conduct interviews to figure out what products are likely to sell. Businesses conduct job interviews to try to find employees. Social scientists conduct interviews for their own somewhat different purposes. But what all these approaches have in common is the attempt to gain information from individuals on some topic. Interviewing is at the heart of social research. If you look through almost any sociological journal, you will find that much social research is based on interviews, either standardized or more in-depth. In fact, some argue that it is the most popular form of data collection in sociology (Den-zin 1989). Examples of interview research abound. William Finlay and 83 James Coverdill ( I WW), lor example, used interviews willi lorporate head hunters to determine who benefits most from the use of such recruiters: the managers who use them or others in the firm, including human re^ source departments. They found that the managers who used headhunted tended to benefit the most, not only because headhunters are able to find prospective employees who are happy in their current jobs (and thus not likely to be looking for work) but also because the managers are able to exercise greater control over the process. In a very different setting, Danti Britton (1999) interviewed correctional officers who worked in men's anil women's state prisons to determine why prison guards prefer to work in men's prisons and what that might say about sex segregation in the labor market. Many researchers combine participant observation with in-depth interviews. (In fact, many textbooks treat the two together.) During the process of observing, researchers naturally ask questions about the ongoing action. Sometimes, these interviews may be informal, as when a researcher spontaneously asks questions of an informant. Other times, they may be more formal, as when a researcher prepares a list of questions. Some researchers use both. Finlay and Coverdill, for example, conducted formal interviews with the headhunters, which they tape-recorded and transcribed. They also did over 300 hours of fieldwork, during which they asked questions more informally (Finlay and Coverdill 1999, p. 14). INTERVIEWING AS A RELATIONSHIP Most methods textbooks treat interviewing as a conversation between two people—the interviewer and the interviewee. The interviewer asks questions, and the interviewee responds to them. But if you think about it, this is a peculiar kind of conversation. In an interview, one person—the interviewee—reveals information about him- or herself; the other does not. One person—the interviewer—directs the conversation, often with expectations for what should happen during that conversation and for what constitutes a "correct" answer; the other does not. One person—the interviewer—decides when the questions have been satisfactorily answered and closes the conversation; the other does not. In this sense, an interview is an odd type of conversation indeed. In what other kind of conversation is there such a lopsided exchange? I prefer to view interviewing as a form of relationship between two individuals (or, in the case of focus groups and group interviews, more than two individuals). The individuals may be close or, perhaps more typically, distant. The interview may be prolonged, repeated over time, or very brief. In each case, however, two individuals come together to try to create meaning about a particular topic. While participating in this relationship, they also draw on established social conventions. For example, questions and answers usually follow one another, with individuals taking turns speaking and observing rules for finishing conversations (Schegloff and Sacks 1974). In most interviews, one person does most of the questioning, with the focus on the person being interviewed. (As we'll see, however, some interviews are far less structured and more like a "real" conversation.) The other characteristic of the interview is that it is focused around the production of talk (DeVault 1999). Thus, while interviews may be a peculiar form of conversation, they still focus on language and its social organization. TYPES OF INTERVIEWS There are several types of interviews, including structured, semistructured, and unstructured. Interviews vary according to the amount of control exerted by the researcher during the interview and to the degree of structure. Structured Interviews At one end of the spectrum are structured interviews, the most formal and the most rigidly controlled type. Structured interviews are more likely to be used in survey research, in telephone interviews, and in market research and political polling. In structured interviews, the sequence of questions and the pace of the interview tend to be preestablished. Although at least some of the questions may be open-ended, allowing interviewees to respond in their own words, they may also be closed-ended, forcing interviewees to choose between fixed responses. In structured interviews, the interviewer usually is not allowed to deviate from a rigid protocol (or interview schedule). The questions must be asked exactly as written, and follow-up questions (also called probes), if they are allowed, are standardized. And if a respondent doesn't understand the question, the interviewer typically does not rephrase it in the respondent's own words. Instead, he or she simply repeats the question, perhaps with minor changes in phrasing. The following extract gives an example of a segment of a structured interview. Notice that the instructions for what the interviewer should say to the respondent are written out. first, I'll like to ask you some questions about youi Inniseliolil.'I'lien I'm going to ask you some questions about your ilaily ai livilies. 1. Besides yourself, how many people usually live in this household? 2. Are there any children under the age of 18? [ If yes] How many? 3. Of all the people living in your home, how many are full-time students? 4. How many people work outside the home, for pay? I'd like to get a sense of what a normal weekday in your household looks like. 5. What time do you typically get up in the morning? In structured interviewing, interviewers typically do not reveal any personal information about themselves, even if asked directly. Rather, they seek to remain as neutral as possible in how they present themselves. Personal revelations on the part of the interviewer are said to produce bias, because interviewees will tend to give the responses that they think the interviewer wants to hear. This is sometimes called social desirability bias. That is, respondents will want to give the response that they think is socially acceptable. So, for example, respondents may underestimate how often they engage in behaviors seen as socially undesirable, such as drinking alcohol or using illegal drugs, or they may overestimate how often they engage in socially desirable behaviors, such as exercising or eating healthy foods or attending church (Presser and Stinson 1998). Researchers who study stigmatized groups or topics that are typically considered private—such as sexual behavior—face additional constraints. If respondents believe the interviewer disapproves of a behavior, they are much less likely to respond honestly. Researchers use structured interviews far more often in quantitative than in qualitative research. While some qualitative researchers may incorporate elements of the structured interview in their research, many reject this type of interviewing for philosophical reasons. Structured interviewing allows the researcher to retain a great deal of control over the interview process. Yet many qualitative scholars believe that structured interviews grant too much control to the interviewer. Because the interviewer controls what questions are asked and how they are worded, he or she can overlook issues that may be more important to the interviewee. In addition, respondents may misunderstand what is being asked, and they may lack opportunities for clarification. The researcher assumes that all interviewees will understand the questions in the same way and that the questions address the interviewees' reality. But the questions may have different meanings for different interviewees, and if the questions are not meaningful to the research pin lit ipanls, the interview data will not he useful. Structured interviews can duis risk missing what's most important to the interviewees. Imagine, for example, that you are askeil in a structured interview what vour favorite flavor of ice cream is. The interviewer gives you a series of < In vices: chocolate, vanilla, peppermint, coffee, or chocolate chip. If your l.ivorite llavor happens to be mango, how will you respond? What if you don't like ice cream at all? What if you really want to talk about sorbet or sherbet or gelatti? If these choices aren't offered, then the interview won't ollect how you really feel. Of course, survey researchers have a number of strategies for avoiding these problems. (If you're interested in finding out more about survey research, see Babbie 1990.) Still, most qualitative researchers choose semistructured or unstructured interviews for the greater depth of insight they give into the lives of their research participants. Semistructured Interviews Semistructured interviews (sometimes called in-depth interviews) are much less rigid than structured interviews. In semistructured interviews, the goal is to explore a topic more openly and to allow interviewees to express their opinions and ideas in their own words. As Michael Quinn Patton (1990) reminds us, we can't observe everything we might want to know. Thus, we interview people to understand what life is like from perspectives other than our own. We try to move beyond our own experiences and ideas and to really understand the other person's point of view. Although the researcher typically begins with some basic ideas about what the interview will cover, the interviewee's responses shape the order and structure of the interview. Each interview is tailored to the research participant. Semistructured interviews thus allow for a much freer exchange between interviewer and interviewee. In semistructured or in-depth interviewing, the researcher needs to listen carefully to the participant's responses and to follow his or her lead. The process resembles a dance, in which one partner (the interviewer) must be carefully attuned to the other's movements. Because the interviews are not prescripted, they can sometimes take surprising turns. Thus, in-depth interviews are particularly useful for exploring a topic in detail or in constructing theory. A number of feminist scholars have argued that these interviews are a particularly good way to study women and other marginalized groups (DeVault 1999; Reinharz 1992). Because women historically have been silenced, they have not always had the opportunity to tell their own stories. In-depth interviews allow them to do so. As I'll discuss in greater detail shortly, reseanheis have diHerein opinions about how much of themselves they should reveal in an interview Some believe that the exchange should be more like a "real" conversation, with interviewer and interviewee both participating in the dialogue (see Reinharz 1992, pp. 32-35). From this perspective, the goal of the interview is to jointly construct meaning on some topic. These researchers, who are more likely to share a postmodernist or critical approach to social research, may not hesitate to present their own opinions and beliefs. They tend to call those whom they are studying their research participants, to emphasize their greater role in shaping the research process. Other researchers, however, feel that the emphasis should remain firmly on the research subjects, with the interviewer playing a much more neutral role. These researchers suggest that the interviewer should tailor his or her presentation of self to the research situation. The interviewer should dress neutrally, so as to "blend in" as much as possible. If an interviewee asks a personal question about the interviewer, the interviewer might politely deflect the question or try to answer it after the respondent has presented his or her own point of view. The question of how much of the self to present in an interview hinges in part on the nature of the research. If the researcher is logging many hours in the field, with interviews a secondary component to observation, the participants are likely to have much greater knowledge of and a more personal relationship with the interviewer. If the interview is a one-time encounter, then the relationship between interviewer and interviewee is likely to be much more impersonal. The researcher's personality also influences the level of self-disclosure. Some researchers are much more reserved, and others are outgoing. These personal qualities have an impact on the kinds of relationships the researcher develops. The following excerpt is from an in-depth interview conducted with an at-home mother. In this interview, notice that the interviewer begins with a general question and follows the respondent's lead. interviewer: Can you talk a little about what it was like for you when you were both working and a new mom? What was that like? respondent: First of all, finding day care was very, very difficult for me. I interviewed a dozen people. Um. Most of them were young, young ladies with new babies themselves. And it was very stressful to think about leaving my son, leaving him to be raised with somebody else. I ended up finding a day care person who was in her mid-50s. She was really a nice woman, and I felt very comfortable with her. And the whole surrounding ... And my husband would drop him off in the morning so I didn't have to deal with that separation thing, which was kind of nice. And then I would pick him up at night, so I got that oppoilunily.The day care woman was really nice; she very rarely called me lor any trivial things. 1 kind of wish she would have—you know, you hear now about the day care people keeping logs, you know, about what they ate or how many times their diaper was changed or whatever. So I don't really have a history from, you know, I went back to work when he was 3 months and I quit when he was 9 months. So there's a whole 6-month gap there. It was hard there, it really was. interviewer: What was the hardest thing for you? respondent: Just the fact that, you know, he would be smiling for the first time without me seeing it. And just the fact that, you know, she would be giving him his first cereal. I don't know. I don't know. It was hard. interviewer: Hard. And how did you make the decision to quit your work? Your paid work. respondent: Well, both my husband and I are in finance, so we kind of did it financially. You know, that was our way to make it seem okay. We sat down and put everything to paper. Where we were, what we could afford to do. And then the decision after that was easy. You know. That's us, though. We do everything with paper and pencil. Okay, how are we going to do this, let's figure this out. But it was more an emotional thing because every day it was so hard to get up. I mean, I was getting up at 5 o'clock in the morning, I would work out, then I would take my shower, get him up, and I was nursing. And I was also expressing milk at work. It was really hard. So it was an emotional thing. And that's kind of what set us to putting it down on paper. Like, how could we do this? What could we do to make this happen? So, you know, we just cut down on a few things, we paid off both our cars, we cut back on all kinds of things. Unstructured Interviews As the label implies, unstructured interviews are the least structured of all. Unlike structured interviews, which tend to be preplanned and may be tape-recorded, unstructured interviews are often conducted in a field setting, in conjunction with an observational study. They tend to be more spontaneous and free-flowing, with topics arising from the situation or behavior at hand. The interviewer typically does not have a set of questions prepared in advance. Instead, questions arise more naturally. For example, if you are conducting a participant observation study in a fast-food restaurant, you might ask questions about the work itself or about the workers' feelings about the job during the course of your observation. Of all interview types, unstructured interviews tend to be the most like "real" conversations. CLOSENESS AND DISTANCE IN INTERVIEW SITUATIONS Survey researchers and authors of structured interview texts suggest that there should be distance between interviewer and interviewee. From this perspective, the interviewer is akin to a mechanical recorder, trained lo extract information in an efficient, detached, yet pleasant manner. To avoid bias, the interviewer needs to reveal as little about him- or herself as possible. Interviewers are (within some constraints) seen as more or less interchangeable. The implication is that any interviewer, if adequately trained, can extract the same information from the interviewee. It's as if there are nuggets of "truth" embedded in the interviewee, and all the interviewer has to do is pick them up using a set of mechanical tools. Yet interviewers—and interviewees—are not interchangeable. They each bring different qualities to the interview, based on who they are and the experiences they've had. Because interviewing is essentially a personal relationship, who the participants are matters. Thus, many feminist researchers have stressed that being similar in crucial ways to their interviewees was important in gaining access to them. For example, Patricia Zavella writes about how she informally made her interviewees—Mexican American women workers—aware of similarities between them and herself (1996, p. 146). By disclosing her own ethnic identity and her status as a working mother with a young child, she hoped that her informants would open up to her as well. Similarly, in her study of Appalachian women and domestic violence, Patricia Gagne emphasized her similarities to the women she was studying by sharing information about her own experiences in an abusive relationship (Tewksbury and Gagne 1997). Presenting the appearance of similarity can also aid in developing rapport. For example, Lauraine Leblanc (2000) studied punk girls in North America. She had been involved with the punk scene prior to becoming a graduate student, and she still retained elements of her past involvement, such as tattoos, hairstyle and hair color, and wardrobe. Her appearance as a punk and her familiarity with the scene (she notes that she was sometimes seen as an "old punk") was crucial to her acceptance by the girls. Must researchers always be similar to their research participants? If this were the case, then we could never know about people different from ourselves, and most research would be about White, professional-class academics. Clearly, then, not all researchers study those who are like themselves—nor should they. But the skills involved in developing relationships across social boundaries are complex. Thus, Carol Stack, a middle-class White academic, writes about how she came lo do research among both urban ami rural African American women, often poor, who were very dif-hrent from herself (Stack 1974, 1996). At one point, she describes a group discussion about her research on men and women who were migrating back to the South. Finally giving her support to Stack's project, one woman declared, "You see here a white woman capable of learning" (Stack 1996, p. 103). A standardized survey, in which all the questions were prearranged and selected, would clearly not have served Stack well in her attempt to transcend boundaries of race and class. From a feminist standpoint, Ann Oakley (1981) argues that interviewers must be willing to risk disclosing personal information and developing real relationships with their research participants. In her research on the transition to motherhood, she found that she simply could not follow the advice of traditionalists in response to her interviewees' questions. In this research project, Oakley interviewed women many times, both before and after giving birth. She was also present at a number of the births—certainly one of the more personal and intimate times in a woman's life. Because of the intimacy that developed over time, she found it artificial, and even impossible, to maintain emotional distance. Traditionalists argue that researchers should try to laugh off requests for personal information in interviews or state that they have never really thought about the question. But, Oakley wondered, how could she laugh off or not respond to questions like "Why is it dangerous to leave a small baby alone in the house?" (Oakley 1981, p. 48) Her interviewees wanted to know if she, herself, was a mother, and she saw no reason to withhold that information. Furthermore, Oakley argues, her own experiences of mothering gave her special insights into her participants' lives and helped shrink the distance between herself and her subjects. As a mother herself, she had experienced many of the same things as her interviewees. And these shared experiences were critical for her research. Because interviews are relationships between people—however artificial they may sometimes feel—interpersonal skills are crucial to being a good interviewer. If the person you are interviewing doesn't trust you or feel comfortable in your presence, then the interview is unlikely to go well. After all, why should someone take the time or effort to interview with you? What's in it for them? Even if participants do agree to an interview, they may not be willing to talk honestly or discuss intimate details about their personal lives if they do not feel some level of trust. This is especially true in attempts to research those who are different from you or those from stigmatized groups. The development ol trust between interviewer and interviewee is ollcn called developing "rapport." Traditional interview texts suggest that you should develop enough rapport to get people to talk to you, but not so much that you actually develop friendships with your participants or disclose too much about yourself. The image is almost one of "tricking" research participants into talking to you. More recently, many scholars, especially feminists, have argued that this image of the interview tends to ignore issues of power in the interview situation (Acker, Barry, and Esseveld 1996; Reinharz 1992; Stacey 1996). Ann Oakley suggests that the traditional focus on extracting information from passive research subjects is "morally indefensible" because it treats those being researched as objects. In the traditional view of interviewing, researchers define the boundaries of the relationship; thus, they do not necessarily keep interviewees' best interests in mind. In fact, the researchers' desire to get research participants to reveal intimate details may directly conflict with the participants' desire to keep information private. Oakley argues that the traditional way of thinking about interviews rationalizes inequality: "What is good for interviewers is not necessarily good for interviewees" (1981, p. 40). In addition, she suggests that the preservation of distance and hierarchy between interviewer and interviewee makes for poor interviews. How can researchers expect intimacy when they are not willing to reciprocate? PREPARING FOR THE INTERVIEW Before beginning an interview, you need to make choices about the interview process. Specifically, you need to address questions like these: ♦ What kind of relationships do you hope to develop with those you are studying? ♦ How intimate a relationship is desirable? ♦ How will you gain access to potential interviewees? ♦ What kinds of information are appropriate to disclose about yourself? ♦ Will you conduct relatively structured or unstructured interviews? ♦ Is the interview process a method for gaining information from research subjects or a process of jointly sharing information and creating meanings? Your decisions about these questions will shape how you prepare lor and < onducl your interviews. Deciding Whom to Interview Methodologists tend to worry a lot about whom to interview, and their concern makes sense. If you don't choose appropriate people to interview, you won't get the information you are looking for. Survey researchers and quantitative social scientists tend to use techniques that will ensure what they call a random sample of the population. These techniques allow them to generalize the results of the study to a larger population. But these kinds of sampling methods are not as useful for qualitative research. In qualitative research, we are more often interested in understanding a particular case in great detail. We want to know a lot about a relatively small number of people. Thus, we tend to sacrifice breadth for depth. We typically don't interview thousands or even hundreds of respondents, as quantitative scholars do. Instead, we might interview only twenty-five or fifty people for our research. Some studies are conducted with even fewer participants. Judith Stacey, for example, studied two kinship networks very intensively over an extended period of time (Stacey 1991). Qualitative researchers usually choose research participants for the specific qualities they can bring to the study. In general, you should choose those interviewees who can give you the greatest possible insight into your topic. Sometimes, after an extended period of field observation, you will recognize that there are several different groups within your field setting. In a fast-food restaurant, for example, the distinction between part-time and full-time workers may be important. If so, you will want to interview people from both groups. Sometimes, you know in advance that you want several different perspectives on a topic. For example, you may want to explore the perspectives of old women and young women on domestic violence, or those of Asian Americans and Latinas. This is sometimes called a purposive strategy, in which you intentionally sample research participants for the specific perspectives they may have. At other times, especially in the beginning stages of a research project, it's helpful to use what is sometimes called snowball sampling or chain referral sampling (Biernacki and Waldorf 1981). In this technique, you begin with an initial interviewee—often, a key informant. Then you ask that person to refer you to friends or acquaintances or others who might be appropriate to interview. In this way, your sample "snowballs." In fact, for "hidden" populations or groups of people who engage in stigmatized behavior, this may be the only way to recruit interviewees. For example, suppose you want lo investigate people who use illegal drugs. ('fiances aie, ii you advertise in a newspaper or bulletin board, you will get Cow (if any) responses. Potential interviewees need to know that you are trustworthy. Thus, having a friend or acquaintance vouch for you may be crucial in gaining access, for certain kinds of research, snowball sampling may be the only way to gain access. If you do use snowball sampling, it is often helpful to begin your recruitment efforts in somewhat different social locations. For example, if yon are studying men who have tested positive for HIV, you might want to gain initial contacts with men from different clinics and social service agencies, in addition to seeking out men who are not affiliated with these agencies at all. One of the risks of snowball sampling is that the participants may be too similar to one another to give you the diverse perspectives you want. Lisa Groger, Pamela Mayberry and Jane Straker (1999) caution that as researchers you need to consider those who refuse to participate in an interview or those to whom you can't gain access. What might you not learn, they ask, because of who would not talk to you? In their research on African American elders and their caregivers, they experienced numerous problems gaining interviewees. There were numerous gatekeepers, some of whom may have had a vested interest in not providing access. Some of the elders may have had particular reasons not to be interviewed. They suggest that qualitative researchers need to think carefully about how the choices they make in soliciting research participants may shape the conclusions they come to. Preparing an Interview Guide In in-depth interviewing, the researcher typically prepares an interview guide to help focus the interview. Unlike a questionnaire or a precoded survey instrument, which provides a rigid order and specific wording for the questions that the interviewer must follow, an interview guide helps the interviewer focus the interview. The interview guide lists the main topics and, typically, the wording of questions that the researcher wants to ask. It also usually includes some ideas about follow-up questions (or probes) But ^tfie researcher does not follow the guide rigidly in conducting the interview. | Rather, she or he adapts the questions during the course of the interview, i changing both the phrasing and the order of the questions.Xhej^ ^n^Irtrasl^add^ the participant's responses."Soine researchers like to create two interview guides: a relatively long one that includes many detailed questions and a'short schematic that summarizes the main topics. The detailed guide serves primarily to help the researcher prepare for the interview; the summary helps him or her remember the key topics during the actual interview. Deciding What Kinds of Quest ions to Ask Not all topics are amenable to interviews. It doesn't make sense to ask interviewees about things they can't answer, and some kinds of questions will make lor more productive interviews than others. For example, it doesn't really make sense to ask native English speakers when they first learned to speak English, because they probably don't remember. If you want to know about what people actually do, rather than what they say they do, you should probably use observation. But you can legitimately ask people questions about the following (Patton 1990, pp. 290-293): * Their experiences or behaviors ♦ Their opinions or values ♦ Their feelings # Their factual knowledge ♦ Their sensory experiences ❖ Their personal background Let's say that you are researching workers in a fast-food restaurant, and you want to ask them about their daily routines. You might ask, "What do you normally do when you begin a shift?" Or you might inquire about a particular experience: "Have you ever been stiffed by a customer? Can you tell me about it?" These questions point t<^events or experiences that research participants have had. "~" You might also be curious about research participants'(qginions or val-, ues. In the case of fast-food workers, you might ask them how they think customers (or the workers themselves) should be treated. For example, you might ask, "Some people use the saying 'The customer is always right.' What do you think about that?" or "What do you think about how workers are treated in this restaurant?" Sometimes, you will want to explore the interviewees\^e//'n£X or emotions about a particular topic or event. In this case, you might ask, "How did you feel when you were yelled at by a customer?" or "What do you like about your work?" Other times, you may be interested ir\ factual information*—what the interviewees know. Thus, you might ask, "Have you heard about proposed legislation to raise the minimum wage?" Be aware that factual questions may not lead to an in-depth discussion. But they can sometimes be used to figure out whether to continue a line of questioning. For example, if you only want to find out what employees have heard about the proposed legislation, you won't want to lonliiuie lliat line ol questioning il lliry haven't heartl ol it. You can also ask questions about tlx\w».v<'.v* what (In- interviewees see, hear, touch, smell, and taste. An appropriate question here might be, "Whal does the kitchen smell like at the end of the shift?"These kinds of questions can be especially useful in setting the scene and trying to figure out what the interviewees' daily reality is, = Finally, you can ask background questions-^questions about, say, family background or social class "or income! You might ask, "How many children do you have? How old are they?" Some background questions are especially tricky—for example, questions about income. Sometimes, interviewers will include a short questionnaire for these kinds of information to avoid having to ask about them face-to-face. Structuring and Ordering Questions In developing the interview guide, it's often useful to brainstorm a\ list of questions and topics that you think might be useful to include. Then you can put them in some kind of logical order. For example, for your research in the fast-food restaurant, you might put questions about how the work is actually done in one section and questions about relations with coworkers in another. Keep in mind, though, that the precise order will depend in large part on the interview itself—it's best to follow the interviewee's lead. But it's also a good idea to at least think in advance about the kinds of questions that seem to go well with other questions and to consider some potential follow-up questions. As a general rule, you should place easier, less threatening questions at the beginning and save more controversial or sensitive questions for the middle or end, once you have developed some rapport and established some trust. When I'm preparing an interview guide, I usually brainstorm a fairly long list of topics and questions. Then I circle all the topics and questions that seem related, make sure the questions and topics don't overlap, and delete redundant questions. Finally, I work on the phrasing of the questions, trying to craft open-ended questions that will help spark discussion, rather than close it down. I often find it helpful to work with a key informant or another researcher at this stage, someone who can help me sort through potential redundancies. I usually try to gather feedback early on and pretest the questions before using them on a larger scale. Box 5.1 shows an excerpt from a brainstorming list for my research on stay-at-home mothers. Notice how the items in Box 5.1 are expanded into an interview guide in Box 5.2. In the actual interviews, I didn't necessarily noxs.i liniinslorming lopics lor an Interview Guide for At-I lome Mothers _ .:,:. ČLué4H&ih4 a four whř Hti Joy (wk* (ike " fřruiťřurě 9fa Joy "wh J$é4 wfaá"MM{':] '-wteté JiJ iaéaí t>&x«ž j,irm7 "-i : -ý&uipore 4é!^ h íJsúÍ^^a. box5.2 Excerpt from an Interview Guide for At-IIome Mothers _ A. Typical Day 1. What does a typical day look like for you? 2. What are the greatest stresses in your day? 3. What are the greatest pleasures in your day? 8. Household Work 4. Tell me about the household work in your family. Who does it? [Probes: Different areas of household work: dishes, cooking, laundry, cleanings car repair, yard work. Does your partner/spouse help out?] C. Transition to Being at Home 5. What kind of work did you do before becoming an at-home mom? 6. What did you like about it? 7. What did you not like about it? 8. What was it like for you to stop working for pay? How did you feel about it? 9. When you first decided to stay at home, how did others react? [Probes: Get reactions of partner/spouse, parents, coworkers, friends.] 10. Did anyone support you in your decision? Who? Á ask the questions in order; nor did I use the exact words listed. As an interviewer, I wanted to follow the interviewee's lead. Still, the list provided guidance for the topics to be covered. Making Questions Open-Ended Remember that your purpose in in-depth interviewing is to explore the research participant's reality. You need to make sure that the questions make sense to your research participants and are phrased in language that is appropriate for them.!The questions should open up discussiorUnot close it down. Patricia Zavella (TBSTfrfSf' example, conducted her'interviews in whatever mix of Spanish and English was most comfortable for her interviewees. If you know that the group you are studying uses specific terms for various things, you should use them. But it also doesn't make sense to take on another group's jargon unreflectively. Think how silly I would sound, for example, if I tried to use teenagers' expressions. Beginning researchers often try to use relatively formal language, which looks good on paper but sounds awkward in speech. In the first draft of an interview guide, for example, one group of students included a question about whether their interviewees had ever "engaged in the use of alcoholic beverages at parties." While technically correct, this question sounded stilted when they asked it. It sounded much more natural when they rephrased it to ask if their respondents "ever drank at parties," clarifying as needed that they were referring to alcoholic beverages ("like wine or beer or hard liquor"). You also need to consider that how you phrase questions will shape interviewees' responses to you. EspeqpJtly:ia^ot-depth interviewing, you need to make sure your questions areopen-ended^Try to create questions that encourage your participants to talk/hfft .shut down. There are a number of tricks you can use to try to keep your interviewees talking. Avoid Dichotomies Try to avoid questions that can easily be answered by a simple "yes" or "no." These questions, called dichotomous questions because there are two possible answers to them, can bring a conversation to a halt because they do not encourage interviewees to continue. Here is an example: INTERVIEWER: Do you feel college drinking is getting out of hand? RESPONDENT: No. Avoid Leading Questions Dichotomous questions can often be leading questions—that is, questions that lead interviewees to give a particular response. You should avoid these questions because they don't encourage tin' interviewee to say what she or be actually thinks. Be especially wary of questions that begin with "Don't you think . . . ?" or "Wouldn't you agiee . . . ?" book at this example: I .eading question: "Don't you think that college drinking is getting out of hand?" You can revise dichotomous and leading questions to make them more open-ended. For example, you can reword the question this way: Revised question: "What do you think about college drinking?" Ask Both General and Specific Questions Asking general questions gives v—-your interviewees the opportunity to reflect on what's most meaningful to them. With a general question, you can focus on those aspects of the topic that are most salient. General questions also allow your interviewees to move at their own pace and indicate that you are interested in what they have to say. So, for example, you might want to ask your interviewee questions like these: General question: "Can you describe a typical day at work for me?" General question: "What do you usually like to do after school?" At the same time, it's often helpful to ask more..specific'questions, ones that require your interviewees to draw on their own experiences. More specific questions can also serve as good follow-up questions. For example, compare these two questions below. The first invites respondents to think more abstractly about college drinking, and the second asks respondents to reflect on their own campus. General question: "What do you think about college drinking?" Specific question: "What do you think about drinking on this campus?" Wheney£X,>possible, try to ask questions that encourage respondents to draw or^ specific experiences/For example, if a respondent has trouble imagining a "typical day," you might ask her or him to describe a specific day: "What about the last day you worked? Can you describe what happened that day?" Take Care with the Question "Why?" Be cautious about using the question "Why?" People don't always know why they do things, and they may feel delensive when asked to provide an account of their ,u lions. This doesn't mean that you should never ask "Why?" in an interview. As a researcher, you are certainly interested in figuring out why people do things or hearing the accounts they give. But it does mean that you should think about whelhei the wording of the question is useful. Consider the difference between these two questions: Potentially threatening: "Why did you get drunk at that party?" Potentially less threatening: "Can you talk a little about your drinking at that party?" The second question might also be followed up with several more specific questions—for example, "How did you begin?" and "Do you remember how you made the decision .ta-keep drinking after you felt tipsy?" Sometimes, simply asking'^'How come?" instead of "Why?" will make a question less threatening. In general, if you can normalize the experience you are asking about and show that you are not being judgmental, you will be less likely to threaten the interviewee. Pretesting Your Interview Guide Youshduld always try to pretest your interview guide, because what may seem like a good question in the abstract may turn out not to be in practice. Sometimes, the language is too formal.or informal, or the question simply doesn't make sense. As a first round of pretesting, you might try the questions on your classmates. Then you might try the questions on a key informant, one who can help you make adjustments. In large-scale interview studies, researchers may do a round of pretesting on a smaller group of subjects before using the interview questions on a large scale. Whether you do a formal or informal pretest, be sure that you are familiar with the questions you want to ask. By the time you interview "real" research participants, you should have the questions basicallyfhiemorized.'|rhat way, you can focus on what the interviewee is saying, not on whether you remember the questions. BEFORE THE INTERVIEW Once you have completed your interview guide and pretested it, you are ready to set up the actual interview. It's best to set up the interview in advance. You should have some idea of how long the interview should take, based on your pretest. Make sure you allow more time than you think you will need lor (he interview. When you .set up the interview, be sure to tell potential interviewees about what you are researching and why. Give them an estimate' of how much of their lime you will need, as well as a phone number or way to contact you if they need to cancel or change plans. If you've made a date for the interview far in advance (say, several weeks), it's a good idea to call the day before to remind them that you'll be coming. And if you are conducting the interview at an unfamiliar place, be sure to j;et directions! Establishing a Location The location of the interview can vary, depending on you and your interviewees' preferences. Interviews can take place in an office, in an interviewee's home or workplace, or in a coffee shop or other public place. Ideally, you should do the interview in a fairly quiet place where you won't be disturbed. If you can, try to turn off cell phones and pagers and other potential sources of interruption. Realistically, however, it's often difficult for some people to find a quiet place. For example, when I was interviewing at-home mothers, I typically came to their homes. Because the women were caring lor their children at home, they tended not to have baby sitters or other child care. The interviews thus were often interrupted so the women could attend to children's needs. While the interviews often felt disjointed and were difficult to transcribe, they also gave much more insight into the texture and quality of the women's lives than an interview in a sterile, artificial environment would have. Constructing a Face Sheet f Often, interviewers construct a face sheet that includes demographic information about the interviewee and information on the set-up of the interview: the name or code number of the interviewee, contact information, the place and time of interview, and so forth. Deciding What to Bring My "interview packet" includes a face sheet, an interview guide, consent forms (and pens), and any other material I think I might need, such as informational handouts for the interviewees. If you are taping the interview, be sure to bring extra batteries and more tapes than you think you might need—and make sure your tape recorder is working. Most interviewers have suffered the unhappy experience of having a tape recorder stop in the middle ol the interview and thereby losing irrcplai cable interviews Some interviewers like to record the date, location of interview, anil name (or code number) of the interviewee at the beginning of the tape. Thai helps them test the recorder, as well as keep track of their interview tapes later on. Dressing Appropriately At the risk of sounding like an old fuddy-duddy, I strongly advise you to think about issues of dress and appearance for the interview. What yon wear, how you style your hair, and how you adorn your body send cleat-messages about who and what you are. And interviewees are no less susceptible to these messages than other people. Thus, you need to consider the kinds of messages you want to send. You'll clearly make different choices depending on your interviewees. If you're interviewing other students, for example, you can probably keep your body piercings in and dress a little more informally. But if you're interviewing residents in a nursing home, you probably should cover up your tattoo and dress more formally. The key is to know your audience. What kinds of assumptions will they make about you? How will they respond to certain kinds of dress or appearance? If you are interviewing students in a sorority or fraternity, wearing a Greek letter pin will give them one impression. (If you don't legitimately have one, though, I wouldn't recommend it.) Wearing a heavy metal tee shirt will give them another. When in doubt, dress a little more conservatively or more neutrally than you think you might need to. DURING THE INTERVIEW Before you actually begin the interview, take time to describe your study and your interviewee's role in it. You should make sure your interviewee knows that she or he can stop the interview at any time or refuse to answer any question. I always inform my interviewees that I'm interested in their viewpoints and that there aren't any right or wrong answers. Some interviewees are not used to structured interviews or are convIncecTthat there's a "right" answer. During the interview, they may ask things like "Is this what you're looking for?" I try to reassure participants that I want to hear about their experiences in their own words. Also make sure that you answer any questions the interviewee might have. Sometimes, for example, interviewees want to know why you have selected them for the interview. If you have a consent form, make sure the interviewee signs it before beginning. ! If 1 am tape rei ording inlei views, I often lind it helpful to move the tape recorder close to the inteiviewces while going over the consent forms and providing preliminary information. I usually reiterate that they can stop the interview at any time, and I show them the "stop" button on the tape leeordcr. I also let them know that they can erase any part of the tape they like. Essentially, I give them control of the tape recorder. Warming Up Interviewees often are nervous about the interview or about being tape-recorded. Many researchers find it helpful to chat for a few minutes before beginning the interview. For example, I asked mothers about their children in my interviews with at-home mothers. Some interviewers like to establish rapport with their interviewees by finding things they have in common. Keeping the Conversation Rolling In-depth interviewing is an art as well as a skill. More than any other research technique, in-depth interviews require activejistening skills. Conducting an in-depth interview entails sending out the message that what your interviewee has to say is important. It's not simply a matter of mechanically asking questions and giving responses. Instead, it involves actively listening to what your interviewee has to say, following up, and keeping the conversation rolling. Think of it as more like a meandering river and less like a game of Ping-Pong. You can encourage conversation through body language and verbal cues that indicate active listening and genuine interest in what your respondent has to say and through follow-up questions and prompts. Some people are naturally better at interviewing (and, more generally, at conversing) than others. Still, interviewing is a skill that you can learn through practice. One way to encourage the flow of conversation is through the careful use of transitions. How can you move the interview smoothly from one topic to the next? Suppose you are interviewing a fast-food worker about her relationships with coworkers, and you want to turn to her relationship with the manager. You might try a transition like this: "You said you tend to get along with your coworkers pretty well. Is that the case for your manager, too?" Sometimes, there won't be a graceful transition. In those cases, it's still a good idea to let your interviewee know where you're heading in the interview—for example, "We've been talking about how you get along with your coworkers. Now I'd like you to think about how you get along with customers. Can you talk a little about that?" Thinking on Your I eel: i ollow-U|>s and Prolx-s In an in-depth interview, you have to "think on your feel." The interview guide gives you a place to begin your questioning, but it's important to remember that your goal is to follow your interviewee's lead. This means that you don't have to go through your questions in lockstep order. Instead, if a question leads the interviewee to talk about something that is farther down on your interview guide (or isn't on the guide but seems relevant), don't hesitate to jump to that topic. Sometimes, interviewees are reticent or simply don't know how much information you are looking for. You can use follow-up questions and probes to clarify responses or to obtain additional information. Some follow-up questions are planned; others are unplanned. You might want to follow up when an interviewee uses an unfamiliar term, or when you need more information to understand a story, or when you simply want more details. The wording of probes and follow-up questions usually flows from the specific topic you are investigating. Still, there are a number of general probes that you might find helpful, including these: * What happened? 4> When did something happen? ♦ Who else was there? 4 Where were you? # How were you involved? * How did that happen? ♦ Where did it happen? ♦ What was that like for you? One follow-up question I use frequently when I want to know more is simply "Can you tell me a little more about it?" As a final question, I usually like to ask respondents if there is anything they want to add or any aspect of their experience that the questions didn't cover. I usually also ask if they would like to ask me anything before I go. Speaking and Keeping Quiet One of the hardest things for many beginning interviewers to do is to deal with periods of silence in the interview. But some people think faster than others. Sometimes, people need time to gather their thoughts and sort through what they want to say or simply to remember something. If you rush in to "fill the gap," you may stop their thought process or cut them off. Thev may feel that yotne mil u ally interested in what they have to say or Ili.it they're going into loo nun li detail. II long silences in conversation make you uncomfortable, you need to lnul some way to stop yourself from rushing in. Sit on your hands. Count silently to yourself If someone really doesn't understand the question, he or she usually will say something or look confused or otherwise indicate a need for clarification. However, don't jump in with a clarification until you're absolutely certain that it's necessary. Look for subtle cues that the person is really finished speaking before moving on to the next question. Sometimes, for example, interviewees might repeat the answer or give a verbal cue that they're done, such as saying "That's all" or "Is that what you mean?" They might look away while they're thinking and then make eye contact when they're ready to speak (or done speaking). If you're not sure if someone has finished speaking, you can always ask if she or he wants to add anything else. Whatever you do, make sure you allow the interviewee lots of time to think. A slower-paced interview will usually give you much richer information than a fast one. In a similar vein, don't rush in to tell your own stories. Sometimes, an interviewee's story will remind you of some of your own experiences, and you may be tempted to jump in. But I strongly recommend that you do not. Instead, keep the focus on the interviewee. Communicating Nonverbally Not all meaning is communicated verbally. A skilled interviewer also watches for body language, which can provide important clues to the respondent's meaning. For example, a fast-food worker might say, "I 1-o-o-o-v-e my boss," and then roll his eyes, indicating the opposite. You need to pay attention to how people say things, as well as to what they say. You also need to monitor your own nonverbal communication. Your body language can send the message that you're interested in what your interviewee has to say—or bored by it. Be an active listener. That is, lean in toward your interviewee, nod your head, and say, "Uh huh." These "minimal prompters" communicate that you are listening intently to what your interviewee has to say. Sometimes^restating what the interviewee has just said is a good check to make sure that you understand—for example, "You said you don't like working the front counter, right?" Keeping on Track What if your interviewee goes "off track"? How much control should you try to exert during the interview with a very talkative interviewee? Most first-time interviewees are too concerned with asking all their preplanned questions in order and keeping interviewees "on tnuk." Hut some ol the most important insights come from the spontaneous parts of interviews. I'm this reason, I recommend that you allow your interviewees a fair amount of leeway. The most important thing is to let participants speak, not to keep TKenTrigidly on track. That said, there are times when an interviewee wan ders very far afield from the topics at hand. When that occurs, it's best to gently guide the conversation back on track. Taping and Taking Notes Should you tape-record the interviews? Take notes? Do both? Some interviewers feel very strongly that you should tape-record all interviews; others feel equally strongly that you should not. In informal, unstructured interviews, such as may occur during participant observation, it may actually be impossible to record the responses. In these cases, taking notes (written or mental) may suffice. In more structured interviews, I recommend tape-recording unless the interviewee seems so unnerved by the process that the interview does not get off the ground. Some interviewees are afraid that they will seem ignorant or say something they'll regret having on tape. Others become painfully self-conscious. Still others—such as those involved in illegal activities—may be concerned that the tape will fall into someone else's hands and be used against them. My experience is that most interviewees eventually forget about the tape recorder. Taping enables you to listen more fully to what the individual is saying. If you are frantically trying to take notes, you cannot make eye contact or give the interviewee full attention. Taping also enables you to go back and listen again and again to the interview—a big help in conducting the analysis. Taping also lets you pay attention to small details or to particular ways of phrasing things that you might otherwise miss. Marjorie DeVault (1999) reminds us that how someone says things may be as important as what they say. The tap_e_recording provides a record of this. Should you take\notes|luring the interview as well? Many interviewers do. Notes can help you remember details about nonverbal gestures or questions of emphasis. They can help you remember where the interview has gone and avoid repeating topics. You can also jot notes about follow-up questions that you want to ask. If you take good notes, then you can reconstruct the interview if the taping fails. I generally do not take many notes during the interview itself, though I keep pen and paper handy. Despite my best intentions to take notes during the interview, I find it hard to concentrate on what people are saying if I'm writing. But others have the opposite .-xpei ience: The act of writing keeps them focused on what is being said. You should use your own judgment about what works lor you. Al lilt THE INTERVIEW Writing Field Notes As mentioned previously, I recommend writing detailed field notes immedi-j alely after an interview.^ often end up taking notes in the car a few blocks away. In those notes, I try to recall as much as I can about the small details of the interview. I try to recall the setting and the appearance of the interviewee, as well as any details about the interaction that strike me. I also try to record my impressions about how the interview went. The following is an excerpt from field notes that I wrote after an interview with an at-home mother: I just came from interviewing 009, who lives in a rented town house in Mill City. She strikes me as depressed. At least more than a little unhappy and isolated. It seems to me that given how much she has moved around and how she hasn't been able to finish her schooling, no wonder she is unhappy.... I was struck by how she talked about her unhappiness in mothering right in front of her daughter. When I got to her house, her daughter was watching television. The house is small, worn, definitely not new. They're saving up for a down payment for a home of their own, which she hopes they'll be able to buy next spring. We sat at her dining room table near sliding doors. She apologized for the view (the next door neighbor's back yard, which had a lot of junk in it). She's definitely not happy with the house—no yard for her daughter to play in, not enough room.... At the end of the interview (and I could see she was nervous about the tape recorder), we talked a little. She said something like "now let's turn the tables. I'll interview you. Tell me about your childhood!" She's uneasy about her mothering, and I felt like I needed to reassure her. I think she thinks that all other mothers are perfectly happy, that other mothers have perfect children, etc. I gave her some info about some of the hard times other mothers have described. Transcribing and Making Sense If you do tape-record the interviews, you should transcribe the tapes as soon as possible after the interview. Don't wait until you have many tapes piled up, or the job of transcribing (and subsequent analysis) will become overwhelming. If you're lucky and have a resean h giant to hire a tninscrlp tionist, you can delegate that work. Hut chances arc, you'll have- to do the transcribing yourself. Transcribing interview tapes is hard work. Even if you are an excellent typist, it still can take many hours to transcribe one inter view tape. A tape player with foot pedals and variable speeds is an enor mous help. Still, according to one estimate, it takes a fast typist about 4 hours to transcribe each hour of interview tape (Morse 1998). A 1 -hour interview can yield twenty-five pages of typed transcript or more. What should you transcribe? How much detail should you go into? How should you record the stuttering, repetitions, silences, laughs, and awkwardness of spoken language? In general, you should transcribe in as much detail as you can muster. Be sure to include both the questions and the responses—what you say, as interviewer, is a crucial part of the conversation. But you need to make a decision about how much detail to go into. Some researchers, called conversation analysts, pay attention to how things are said (Schegloff and Sacks 1974). They are interested in the rules that structure conversation, and they have developed very complicated methods for transcribing tapes, including symbols to indicate pauses between turns, intonation, and overlapping speech. For your purposes, you probably won't need to go into that much detail. Still, it's useful to reproduce the speech as faithfully as you can. I don't recommend "cleaning up" the speech too much (for example, don't correct grammar or transform spoken language into standard written English). Spoken language, after all, is much less formal than written language, and it's important to preserve the flavor of what was said. Some interviewers give a copy of the tape or transcript to the interviewees afterwards to let the interviewees make any corrections or additions. For example, Sharon Thompson interviewed over 400 teenage girls over a 9-year period on the subject of romance and sex (Thompson 1995). After each interview, the participant could decide whether to allow Thompson to include the tape in her research. The girls were encouraged to call her later if they changed their mind. group interviews The Role of the Focus Group Not all interviews are conducted with only one person at a time. Some researchers conduct interviews with small groups, typically with fewer than - !^i;en PeoP'e- Sometimes called focus groups, small-group intervi^w71Tr^used~ extensively by market researchers and political pollsters. Although the tech- niques lor locus groups were originally developed in the 1940s and 1950s, iIm-v subsequently fell out of favor among most social researchers (Morgan I'iMH). In the past two decades, however, the use of focus groups by social i. .can hers has become much more commonplace. Social scientists often use focus groups to evaluate programs (such as lob training programs or drug and alcohollieatment programs). Public health re-searchers, and individuals interested in "social marketing," or trying io introduce desired behaviors such as sexual practices that decrease the risk ..I IIIV, also use focus groups (Morgan 1996). Sometimes, focus groupsjire iced alone; but more oftenfthey are used in conjunction.With other meth^) ...l.s, such asindividual interviews or surveysySocial scientists have used locus groups to explore a variety of topics, including fear of crime among I alinas (Madriz 1998), people's experiences with natural disasters such as I lurricane Andrew (Belgrave and Smith 1995), the coping strategies of I MV-positive mothers (Marcenko and Samost 1999), and women's feelings about sexuality (Montell 1999). Group interviews, like individual interviews, can be relatively structured or unstructured/They are useful when you want to know about peo-pie's^"pTnTor^r'attitudls^ather than people's actual behavior. They are also useful when~you want tofun3erstand group processes^bow people arrive at decisions. When they Wr4rWeIl~T6cus groups can be an extraordinarily rich source of data, as focus group members build on one another's ieleas and opinions. One of the advantages of small-group interviews is that they allow for the collection of a fairly large amount of data in a relatively short period of time. With group interviews, you can typically sample a larger variety of eipinions in a shorter period than in individual interviews. Overall, they can be less time-consuming (and thus cheaper) than individual interviews. In addition, some researchers find that focus groups are especially helpful for studying transientjiojmladon^ For example, if you want to conduct interviews witlTlnigrant farmworkers, who move from place to place picking crops, it will take you a long time to conduct individual interviews. By the time you are partway through, the workers likely will have moved on. Focus groups provide you with a way to gather interview data more quickly. Some feminist researchers argue that focus groups are especially helpful in( reducing the imbalances of .geavet-between the researcher and those bemg"res7arched (Montell 1999). By enabling women to speak with others who have had similar experiences, focus groups help empower women. Esther Madriz sees focus groups as a form of "collective testimony," a way for women to break their silence and confirm their experiences with other women (Madriz 1998, 2000). A focus group typically consists ol a small number ol participants usu ally less than 10. Often, the focus group is relatively homogeneous with regard to age, ethnicity, gender, or some other characteristic important lu the study. What's most important in forming a focus group is liiuling a group of people who will feel comfortable interacting with one another and who will express their opinions freely. Sometimes, groups are composed ul people who are already familiar with one another, such as parents in a par-ticular school or workers from the same workplace. Other times, the participants are unfamiliar with one another. Recruitment can be a difficult problem for focus group researchers (Morgan 1995). Unless the topic is of great intrinsic interest to potential participants, they may not be motivated to participate. And once they have agreed to participate, there is still the possibility that they will not show up. Yet, unless at least a minimal number show up, the focus group won't he as productive and may even have to be canceled. Thus, experienced users I of focus groups often use extensive follow-up procedures (written anil i telephoned reminders, for example), offer incentives (food or money), and | typically over-recruit. One rule of thumb is to invite at least two more par-I ticipants than you think you will need (Morgan 1995). The Role of Moderator Focus groups can be more or less structured. Morgan (1996) argues that it's helpful for moderators to control the session.. in„lerjms~©£ two dimensions. First, moderators may play a more orjess directive role}in regard to the questions asked. In a more structured focus group, the questions are relatively specific, and the moderator controls what topics are brought up. In this group, the moderator may move group discussion away from topics that are not of interest to the researcher. In a less structured group, the questions can be very general, and the moderator allows topics to emerge according to the group's interests. The second dimension of structure involves( group dynamics?yln a relatively structured group, the moderator plays a more "HirectiveroTe in terms of group dynamics. The moderator can intervene to encourage quiet or shy participants and at the same time tactfully discourage more forceful individuals from dominating the group. Some experts suggest that it's helpful to have two moderators: one to focus on group processes and keep the conversation moving and the other to observe and take notes. Because each group member must be^allewedLjtime to participate, a group interview will typically deal with^ěwer topics);than an individual interview. Most focus groups last betweenfl and 2 nours?»In that period, only a relatively small number ul Inpu s i)an be introduced. Sustaining dis-> ussinn is easier if the participants are legitimately interested in the topic. Some experts argue thai it can be helpful to begin with questions that are likely to be ol most interest to participants, even when those topics. areriU. i he most important to the researcher (Morgan 1995). Stimulating free and open discussion is one of the greatest challenges of i he moderator. One risk of focus groups—especially when they include individuals of differing power and status or when they focus on controversial topics—is that participants might censor themselves and defer to group opinion so as not to "rock the boat" or make others feel uneasy. This phenomenon, ca\\e4^groupthink^f Irving Janis (1982), can be avoided. One way is through the use of "devil's advocates"—individuals whose specific role is lo question the group's ideas and decisions and thus to spark free discussion (MacDougall and Baum 1997). Another possibility is to include only those who are of similar statusthe group, thereby avoiding the problem of lower-status group members censoring themselves. Esther Madriz argues that it may be important for the moderator to be of the same racial/ethnic background as the participants (1998). Sharing important similarities, she suggests, leads to rapport, which enhances the likelihood of a successful locus group session. Ethical Issues in Conducting Focus Groups Focus groups have some specific ethical issues. Unlike in individual interviews, the researcher is not the only one who needs to respect confidentiality. In the case of group interviews, all the other group participants need to maintain confidentiality as well. Participants won't speak freely if they believe that what they say will not be held in confidence. This is especially important in focus groups that deal with sensitive issues, such as sexuality. Bruce Berg (2001) suggests that all participants in a focus group sign a confidentiality statement in which they promise not to reveal information outside of thegroup. Others sujgest^beginning the focus group with a statement of the ground rules for participation^ncluding the confidentiality pledge. --------- -----■* , n xl ^ ■J 4o ,/H-C Recording the Focus Group How should you keep a record of the conversation? Group conversations are complex: People sometimes interrupt or talk over one another, and the conversation may move rapidly from one side of the room to the other. Audiotapes are useful, but they do not indicatejwhcij^espeaker is—something that can cause problems in the analysis. ^Vritten notes^are usually necessary to augment audiolapcd locus groups. Sonic locus group lacililators video tape the session. (Market researchers, who are typically far belter funded than academic researchers, may have access to plush video conference room', with one-way mirrors so that researchers can view the session. If you have access to resources like these, of course you should use them!) Whatevei method of recording you use, it's useful to keep written notes as well, and tapes (just like individual interview tapes) should be transcribed as soon as possible. MAKING MEANING What do you do when you're finished transcribing the tapes? How do you begin to make sense of the data you are collecting? We will deal with strategies for analysis in detail later in the book, especially in Chapter 8. In the meantime, you should 'listen to the tapes and read over your transcripts several times. Try to listen closely to what your research participants have T)eon saying. It is only through 'immersing yourself in their words that you will begin to make sense of the conversation you have been a part of. QUESTIONS FOR THOUGHT 1. How much of your own opinions do you think you should share in an interview situation? What might be some of the consequences of sharing personal information with an interviewee? What might be some of the consequences of not sharing? 2. What kinds of topics might be better suited to an individual interview than to a group interview? What kinds of topics might be better suited to a group interview? Why? 3. How do you see yourself? Are you extroverted and relatively outgoing? Or are you shy and introspective? What personal challenges do you think you will face in trying to conduct interviews? What strengths do you think you have? EXERCISES 1. The purpose of this exercise is to develop your listening skills. You'll need to do this one with a friend. Have your friend talk for 3 minutes (use a Iiinet ) on any lopii that nileie'.ls her or him. Your job is to listen as i arelully as you can without interrupting. When the time is up, try to I'eeall as much detail as possible about what your friend said. ' ( )bservc someone who is actively listening to another person. Pay attention to the body language that they use: leaning in, shaking their head, and so forth. What kinds of nonverbal cues tell you that someone is listening? What kinds of nonverbal cues tell you that someone is not paying attention? Develop an interview guide on a topic you are interested in. When you have completed an initial draft, go through the questions looking for dichotomous questions and leading questions. If you find any, rewrite them. ■1. Try out your interview guide on a friend or fellow student. As you're interviewing, pay attention to which questions seem to stimulate discussion and which seem to shut it down. When you have finished, revise your interview guide based on your experiences. It's often helpful to have a third person act as an observer and give you specific feedback on what went well and what could be improved. 5. Develop a set of questions for a focus group that you could conduct with a group of fellow students. Campus life is full of potential topics. After you have developed the questions, try conducting a focus group with your classmates as participants. (i. Imagine that you are interested in studying youth violence. Develop a strategy for recruiting participants for either a focus group or an individual interview. (Alternatively, develop recruitment strategies for studies of police harassment, worker safety in small businesses, or parents who homeschool their children.) SUGGESTIONS FOR FURTHER READING Fontana, Andrea, and James H. Frey. "Interviewing: The Art of Science." Pp. 47-78 in Norman K. Denzin and Yvonna Lincoln (eds.), Collecting and Interpreting Qualitative Materials. Thousand Oaks, CA: Sage, 1998. Holstein, James A., and Jaber F. Gubrium. The Active Interview. Thousand Oaks, CA: Sage, 1995. Madriz, Esther. "Focus Groups in Feminist Research." Pp. 835-850 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage, 2000. Morgan, David 1.. Focus Gm///« „, Qualitative Research. N.-wUiy l\nk, (:A: Sag,-, 1988. ' ' Oakley, Ann. "Interviewing Women: A Contradiction in Terms." 1'p. 3(M,'I in />„,„, Feminist Research, edited by Helen Roberts. London: Routlcdgc & Kcgan l\,ul,' Seidman, Irving. Interviewingas Qualitative Research: A Guide for Researcher, in /■',/„ canon and the Sotial Sciences, 2nd ed. New York: Teachers College Press, l!)!IH. Unobtrusive Measures Analyzing Texts and Material Artifacts e average person throws away 2.1 pounds of garbage a day, much of it paper and plastic (Rathje and Murphy 1992). What would people find if they systematically rooted through your garbage? No doubt lots of paper, some food wrappings, some leftover food, and maybe some worn-out clothes or empty printer cartridges. How much evidence of fast-food consumption would they find? What about fresh vegetables, or milk, or soda? What might they be able to tell about your study habits? Your work habits? Your choices in recreation? That's exactly what William Rathje, an archaeologist and director of the Garbage Project at the University of Arizona, does. Project researchers systematically identify and analyze what households in Tucson, Arizona, throw away. They also examine landfills, sometimes digging down deep to analyze trash from earlier decades. By systematically analyzing household garbage, they have learned a great deal about people's habits and behaviors—especially those that people don't necessarily want to talk about, like drinking alcohol or eating processed foods (Rathje 1992, 1993). They have even devised ways to estimate the size and composition of the population based on garbage. For example, they can fairly reliably tell how many babies live in a given neighborhood by the number of diapers they find in the trash. The work of the Garbage Project is a good example of using unobtrusive measures to study human behavior. Unobtrusive measures involve any form of studying human behavior that does not rely on asking people directly (such as interviewing) or on observing people (such as doing 115 PHYSICAL TRACES One way in which you can study people without interacting directly is through the study of physical traces (Webb, Campbell, Schwartz, Sechrest, and Grove 1981). Humans leave physical traces of their various activities behind as they go about their daily lives, and you can use this evidence to make inferences about them. In my neighborhood, for example, you can buy a Chinese-language newspaper from a box on the street alongside the local English-language paper; you can also buy national newspapers in English, like USA Today and The New York Times. This fact tells us that there is a large enough population of Chinese-speaking people in the community to support a Chinese newspaper. In the same neighborhood, you can see many Italian groceries and bakeries alongside Vietnamese noodle shops. Again, this gives us some information about the neighborhood and the ethnic groups living there. A cemetery in a neighboring town shows a similar pattern. Most of the tombstones in the oldest part of the cemetery, dating from the late 1700s, contain English names. In a somewhat newer part of the cemetery, you can read Irish and Italian names on the gravestones. In the newest part of the cemetery are a few Asian, Armenian, and Portuguese names, along with large numbers of Italian and Irish names. A careful social observer can learn something about the migration patterns of ethnic groups by looking at the environment. Generally, physical traces are either measures of accretion or measures of erosion. Measures of accretion entail the accumulation of layers. The Garbage Project is a good example of this. Garbage builds up in layers as people throw things out. By studying the layers, you can get information about how a society or group of people within it change over time. In the top layers, for example, you might see computer monitors and other parts of discarded personal computers in landfills. You certainly wouldn't find evidence of these if you dug down a little deeper; instead, you might find construction debris, bottles, and cans (as you would still find today). If you looked carefully, you might find pull-top tabs from beer and soda cans from the 1960s or eight-track tape players from the 1970s. Another good example of accretion is graffiti (Lee 2000), which can include "official" signs and notices such as stop signs or no-parking signs. These signs, some argue, serve as a continual reminder of the state's regula- tion ol sniial lile. Sometimes, gi.illiti can be more political in nature or express pressing social comerus. lor example, slop signs have been spray-painted to read "Stop Rape" or "Stop War." Recently, concerns about global-i/alion have also been expressed in the form of graffiti. Graffiti can also serve to demarcate social boundaries, as in gang graffiti. Public art, including murals, can also be considered a kind of graffiti. And all of these physical t races can be analyzed. Measures of erosion entail studying how people gradually wear down lacets of their environment through daily use. By studying patterns of wear, you can gain information about people's activities. For example, a landscape architect might place a walkway in an area where people don't habitually walk, but the paths worn in the lawn indicate where people actually walk. In a library, the most frequently used periodicals might be heavily thumbed and worn looking. In a museum, footprints might be worn into the floor of the most visited exhibits, whereas the floors near little-viewed sites might be clean and shiny. Your imagination is the only limit in devising ways to study people's physical traces. In general, this evidence can help you figure out what people actually do, rather than what they say they do. Because this method does not entail interacting directly with people, it also enables greater anonymity for the researcher (Webb et al. 1981). In addition, the biases of this research method are likely to be very different from those of other methods, such as participant observation and interviewing. For this reason, studying physical traces can be a very useful addition to more traditional sociological methods. But this method has drawbacks. Most important, we need to ask, What is the likelihood that any given object will actually survive? For example, vegetables decompose far more quickly than bones. Thus, unless you sift through garbage quickly, you will have more difficulty gaining evidence of vegetable eating than of meat eating. In addition, some physical accretion and erosion may be so small as to be unmeasurable. Walking on a soft surface, like a lawn, will yield more traces than walking on a hard surface, like concrete. You might not be able to measure the erosion of concrete in a short span of time. Finally, because these methods tend not be very systematic, they often are paired with other, more systematic methods. MATERIAL ARTIFACTS One form of studying physical traces is the study of material artifacts, the objects that people produce, like pots and pans and cars and computers. Archaeologists and historians may be more accustomed to studying material artifacts, but sociologists can use them as well to help make sense of the social world. Social scientists have studied tattoos (Sanders 1989), gral'lill (Cole 1991], and Zippo cigarette lighters collected in Vietnam (Walters 1997) , among many others. Analyzing material artifacts can be far more complicated than analyzing written texts. First, material artifacts are less logical than formal language, in the sense that they don't have a formal grammar. You can't make a comprehensive dictionary of all the meanings that an artifact might possess. Furthermore, the meanings of material objects often remain implicit (Hodder 1998, p. 117). Thus, we need to understand the social contexts in which material artifacts are made, used, discarded, and reused (Hodder 1989, 1998) . This means that the act of interpreting material culture has to be tentative. It expresses as much about the analyst—about who you are—as the material culture itself. Second, the meanings of material artifacts may change over time, as may the contexts in which they are produced. When automobiles were first produced, for example, only the wealthy could afford them. Car ownership takes on very different meanings when most people can afford to own one. Third, material artifacts are harder to "read" than a written text because they are not linear. In reading a set of field notes or a transcript of an interview or any other written document, you know to read from left to right and from top to bottom. (Of course, the conventions for reading vary from one language or culture to another. If you were reading a document in Japanese or Hebrew, you would move from right to left, from what feels to an English speaker like the back to the front.) With material artifacts, there is no necessary order to your analysis. Imagine that you are interested in contemporary fashion and want to understand the meanings of a pile of clothes. There is no necessary order in which you should "read" those clothes. From top to bottom? Left to right? Back to front? Analyzing material artifacts is not a linear process. In addition, in analyzing material artifacts, you have to figure out what things mean without having access to the person (or people) who produced them or used them. This is especially true if you are working with historical artifacts. Thus, you may not have an "insider's" perspective on the meanings of things and have to interpret them using contextual clues. Even if you do have an insider's perspective, people are often not articulate about why they do things. Think about it: Why is an exam book usually blue, and why do we still call it a "blue book" even if it is not blue? Why do we usually use a mug for hot drinks like coffee or tea, and a cup or glass for cold drinks? Why not use the same container for both? Having the advice of an insider—one who uses blue books or coffee mugs or glasses—probably won't help you much. Interpreting material arlihuls thus involves a Iwnncnculiail process, a pioeess of making meaning. You need to try to situate the artifact both in the >. mlext in which it was made ami produced and in the context in which you (as analyst) are situated. You need to try to understand not only how the item was produced but also how it came to your attention for analysis. Why was this particular artifact (or type of artifact) preserved, and not others? At one level, this is a physical process. Some kinds of things (like vegetable matter) decompose relatively quickly; others (like ceramic or bone or stone) do not. Yet the preservation of things is also a social process, one that involves human choices about what should be preserved and what should not. What things are seen as worth keeping? For example, you probably don't think twice about throwing out or recycling a Coke can or a bottle that water came in. In another culture, in which material goods are less plentiful, these items would be carefully preserved and reused for a different purpose. How do you go about analyzing material artifacts? It is certainly not easy, nor are there any general guidelines. Analyzing these artifacts always involves the study of local, particular cultures. One analyst states, "There is nothing easy in our work, but its basic strategy is not hard to state. We hunt lor patterns" (Glassie 1991, p. 255). Archaeologist Ian Hodder suggests that the analyst of material artifacts must work "between past and present or between different examples of material culture, making analogies between them" (1998, p. 121). To do this, you need to immerse yourself in the specific historical context (Tuchman 1998), and you need to know enough about the cultural context in which something has meaning. Specifically, you need to address these issues: ♦ Where and when was this item (or group of items) produced? ♦ What do you know about the society in which it was produced? ♦ What kinds of people made the object you are analyzing? How? Who used it? ♦ How was it sold or distributed? ♦ Did different people use the object in different kinds of ways? How? ♦ How was the meaning of the object transformed in different contexts? ♦ What kinds of personal meanings might the objects have had, as well as larger social meanings? ♦ What kinds of things were used in similar ways? ♦ What are the economic and political systems in which the items were produced? These are just some of the questions you might ask to get you started. You also need to consider what historical theories might help you understand the material artifacts you are trying to analyze (Hoddcr I99H). How have different theorists thought about the time period or the type of material culture you are interested in analyzing? It's important to immerse yourself in what others have said about the historical period (whether it is an earlier period or today) and the social context. Finally, you should try to confirm the account you have made. That is, you should ensure that your account is internally coherent, logical, and plausible. You also need to consider any potential alternative arguments. And you need to ask whether your account leads you to understand similar objects and opens up new lines of questioning. Ian Walters' research into Zippo cigarette lighters provides a good example of the analysis of material culture (1997). Zippo lighters were issued to American and Australian military personnel during the Vietnam War (or, as the Vietnamese call it, the American War in Vietnam). Many service personnel engraved the lighters with their date of tour, location of service, or the year. Others engraved them with short sayings or quips. Many thousands of these lighters survived. Today, there is a thriving market in Zippo lighters in Vietnam's tourist markets, and people buy them as war memorabilia. Interestingly, most of the lighters available in Vietnam are fakes—not the actual Zippo lighters from the war era. Yet the sayings engraved on them replicate the engravings of the service personnel. Thus, these lighters have retained meanings far broader than their simple function: to light cigarettes. These lighters can be interpreted in multiple lights. As Walters argues, "As personalized icons, signifiers, they still speak to us long after their original owners have ceased to be fighters in the war zone" (1997, p. 64). In fact, the "life history" of the Zippo, Walters suggests, contains several stages. The lighter began life as a commodity—something to be bought or sold. It was sold to the army, issued or sold to individual soldiers, and used as a lighter (and as a war weapon). With the engraving, the lighter became a sign of unit identity or a personal statement. When it passed out of the owner's possession, it became either junk or someone else's souvenir (or perhaps someone's functioning cigarette lighter). In contemporary Vietnam, it reentered the commodity market as something else to be bought and sold. But it also became a memento (perhaps for soldiers who had served in the war and who returned decades later to buy the souvenir, or perhaps for other tourists), a museum piece (for those Zippos shown to be "actual" war-era items rather than fakes), an item for research, and even an object for discussion in a research methods textbook. Walters argues, "Vietnam Zippos tell us stories. They become speech, lei they mean something. Vietnam Zippos give us insights into the soldiers who annotated them and used them, loved them, laughed at them, and lost them" (Walters 1997, p. 73). But the lighters also tell stories about modern Vietnam and the world economy. To understand the meanings of the Zippo lighter, we need to immerse ourselves in the context of both the American War in Vietnam and contemporary Vietnamese life. Analyzing material artifacts is clearly difficult—especially for those of us who are used to working with written texts. Because the analysis is so contextual, there are no simple rules that you can apply. You need to learn about the culture and history of the objects you want to analyze. But it is very worthwhile work, for it can lead you to understand human behavior in greater depth than can studying words alone. DOCUMENTS AND RECORDS Another way you can study human behavior unobtrusively is through written texts in the form of documents and records. Documents and records are any written materials that people leave behind. These might include things like private letters and diaries, corporate records, and government documents. These might also include media accounts, such as television programs, newspapers, and magazines. In more recent years, they might also include electronic texts, such as e-mail lists and Web sites. Although the different kinds of texts might involve somewhat different technologies, we can study them using similar methods. This chapter focuses especially on methods for obtaining written documents and records and on some of the associated problems in doing so. Chapter 8 examines methods for analyzing texts and making sense of documents and records. (The procedures for analyzing texts are typically called content analysis.) Public Records Many researchers distinguish between two kinds of written texts: documents and public records (Lincoln and Guba 1985; Macdonald and Tipton 1993). Public records include those materials produced for "official purposes" by social institutions like governments, schools, and hospitals. They document official transactions, like births or marriages or sales of houses. While many public records are numeric (a good example of this is U.S. Census Bureau materials) and are often used by economists and other quantitative social IK 'lulling things hi mil tra dm i i|ne of the main challenges in using historical materials is evaluating them. Perhaps more than any other type of data, you need to evaluate historical materials carefully. When you are using secondary sources, you need to be alert to historians' implicit theories and paradigms. History is never simply a iccitation of "just the facts." Rather, historians decide what to report, what details to emphasize, and how to interpret events. In the course of their research, historians typically gather thousands of pages of documents. Obviously, they don't simply print these documents verbatim and let the reader decide for her- or himself what they mean. Rather, they compress the huge volume of pages into a more manageable form. How do historians accomplish this? How do they decide what to focus on and what to ignore? They don't usually specify this process in their published works. As a careful reader, you need to try to figure out what the biases and predilections of the historian are. You also need to figure out if the author's viewpoint is well established. Do other historians say similar things? What kinds of evidence do they use? Does their evidence seem credible? Primary sources require extra care. You need to determine first if the source is authentic. That is, was it actually written by the person said to have written it? When was it written? Where? Again, is the document authentic, or merely a clever fake? You also need to determine if the primary source is representative. As we discussed in the section on material artifacts, not all documents survive. In evaluating primary source documents, it's important to consider which people are more likely to have left primary source documents behind. Literate people, for example, obviously are more likely to leave diaries and correspondence than those who cannot write or read. Thus, because African American slaves were forbidden to learn to read and write, they left few letters and other written materials that would document their daily lives. Much of what we know about African American life during slavery comes from oral histories and the testimonies of former slaves collected after the Civil War. In general, ordinary people tend not to think that their lives are particularly interesting or worth documenting. Ordinary working-class people don't usually donate their private papers to archives or otherwise make their papers accessible to historical researchers. For example, have you donated your papers and personal effects to an archive? This fact leads to an inevitable bias in the kinds of personal materials available lteeau.se people such as presidents or famous actors tend to assume dial their lives are vvoi th documenting, they are much more likely to preserve their papers. Yet nur.i of us don't live like presidents or movie stars. The kinds of historical male rials available tend to make it easier to gain knowledge of famous peoples lives than ordinary people's. In evaluating primary source materials, you need to ask whose perspei tive they reflect and what kinds of interests the writer had. For example, was the author of a slavery-era document a slaveholder? Someone active in the movement for the abolition of slavery? Each position carries with it a per spective and set of vested interests. A slaveholder who emphasized the good living conditions of slaves would probably not be believable. But if an aboli tionist remarked on the quality of slave housing in a particular location, that person might be more credible. Historical records often reflect the interests of those in power. Thus, you need to evaluate materials carefully for bias. You also need to consider how the document was preserved and how it made its way to an archive (or to your hands through some other means). Do you have access to the whole document, or only a fragment? What didn't survive? Did anyone have an interest in suppressing certain parts of the material? For example, the niece of poet Emily Dickinson censored passionate letters that Dickinson wrote to her sister-in-law, Sue Miller (Miller 1995). Censorship like this has made the study of lesbian and gay history (and, in general, the historical study of sexuality) difficult. But not all gaps (or "missing data") result from censorship. Sometimes, materials are simply lost or discarded. Other times, despite people's best efforts to save them, materials may be destroyed. This means that historical researchers are inevitably working with fragmentary evidence. Because the "complete" record never survives, the conclusions you make based on such evidence must also be tentative. Creating Meaning When evaluating historical materials, you also should pay careful attention to meanings, at both a literal (or surface) level and a deeper one. At the surface level, you need to ask what the document means and what language was used. Sometimes, words change from one period to the next, so you need to be certain that you can understand the literal meaning. At a deeper level, you need to know something about the conditions under which the document was produced. Again, this involves knowledge of the specific context and time period. Thus, you need to determine what kind of language was common for the time and place and person who cre- .ited the doi unient. And you need to deteimine whether other documents •.uppoil the perspective or whethei the document seems to be at odds with what others lend to say. I )oing historical and documentary research can be hard work. Perhaps more than any other type of research, it demands knowledge of the particular, local contexts in which various materials (written and otherwise) are produced, consumed, kept, and discarded. It demands a critical eye and an ability to scrutinize sources carefully. Yet, because of their very specificity, historical methods can provide a richer and more finely nuanced study than many other methods. QUESTION FOR THOUGHT What are some of the differences involved in analyzing material artifacts and written documents? What different kinds of information might they give you? Which would you prefer to study? Why? EXERCISES 1. Look around your own neighborhood. What physical traces might help you figure out what kinds of people live there? (If you're having trouble getting started, try checking cemeteries, names on mailboxes, types of stores and restaurants, and so forth.) 2. How might you study the following things without either interviewing or directly observing people? a. Which toys are most popular among children b. Whether college students in a particular school exercise frequently c. Whether people in a particular neighborhood eat healthy foods 3. Locate some material artifacts that you can analyze. (If you're having trouble thinking of something, remember that material culture is all around you. It can be, literally, just about anything.) The items can be contemporary or historical. Try to find out as much as you can about how the things were produced and for whom. See if you can find out some of the meanings these items may have. 4. Choose a magazine that interests you. Try to step back and look at the magazine as a social analyst, not as a consumer. Then see what you can determine about the audience for the magazine. What does the magazine seem to assume are the interests and preoccupations of its readers? >INl What can you tell ahoul the biases ami interests ol iliusc who pul the magazine? (Alternatively, try this exercise using a newspaper, a TV show, or a radio program. If you are choosing a TV show or radio pm gram, you will probably need to watch/listen a number of times.) 5. Investigate what historical materials are available, either in your school library, in a public library near you, in a local historical society, or on I In* Internet. 6. Choose a Web site of interest to you, and analyze its potential uses I research, using the guidelines for evaluating Web sites in Chapter 2. 01 7 Action Research SUGGESTIONS FOR FURTHER READING Hodder, Ian. "The Interpretation of Documents and Material Culture." In Norman K. Denzin and Yvonna S. Lincoln (eds.), Collecting and Interpreting Qualitative Materials. Thousand Oaks, CA: Sage, 1998. Lee, Raymond M. Unobtrusive Methods in Social Research. Buckingham, England: Open University Press, 2000. Neuman, W. Lawrence. Social Research Methods: Qualitative and Quantitative Approaches, 2nd ed. Boston: Allyn & Bacon, 1994. Devotes a full chapter (16) to historical-comparative research. Ragin, Charles C. Constructing Social Research. Thousand Oaks, CA: Pine Forge Press, 1994. Gives more on comparative research. Tuchman, Gaye. "Historical Social Science: Methodologies, Methods, and Meanings," In Norman K. Denzin and Yvonna S. Lincoln (eds.), Strategies of Qualitative Inquiry. Thousand Oaks, CA: Sage, 1998. Walters, Ian. "Vietnam Zippos." Journal of Material Culture 2(1): 61-75. 1997. A good example of analysis of material culture. Webb, Eugene T; Donald T. Campbell; Richard D. Swartz; and Lee Sechrest. Unobtrusive Measures. Thousand Oaks, CA: Sage, 1999. ( kganizing is the active unearthing of people's individual stories, the collective examination of the meaning of those stories in light of our shared story, and the opportunity to write new endings to both our individual and collective stories. -Larry McNeil (1995) Wh can conduct research? Can only people with Ph.D.'s or other specialists conduct research? Or can ordinary people do it themselves? What should be the relationship between the researcher and those being researched? And what is the purpose of research? To enlighten? To entertain? To create "basic" knowledge that has no particular application (at least not yet)? Or should research end with some form of action or improvement in the lives of people? THE PURPOSE OF ACTION RESEARCH Action researchers argue that research should not be aimed solely at creating esoteric knowledge, nor should it be conducted only by people with advanced degrees. The outcome of research should be useful, aimed at improving the lives of those who are the subject of research. Research, says Ernest Stringer, should be "organized and conducted in ways that are conducive to the formation of community—the 'common unity' of all participants—and that strengthen the democratic, equitable, liberating, and life-enhancing qualities of social life" (1996, p. 25). Because individuals are the experts on their own lives, action research should involve community 135 members at all levels, including the development ol'the problem anil strut egies for researching it. M. Brinton Lykes (1997), for example, engaged in participatory action research over an extended period in Guatemala, a nation plagued by war I'm decades. She was invited by a community-based women's association to work with members to respond to and ameliorate the effects of war. One of the projects the women planned was to distribute cameras and film so thai women in the community could tell one another their stories of survival. By documenting their lives on film and meeting to discuss the common themes, the women could begin to research the strengths that they had developed and use the information to improve their ongoing programs for children. Brinton's research is just one example of action research. Others have studied many different settings, including business organizations like Xerox Corporation (Greenwood, Whyte, and Harkavy 1993), Asian nongovernmental organizations (Brown 1993), worker cooperatives in Spain (Greenwood, Whyte, and Harkavy 1993), and older Aboriginal women in Canada (Dickson 2000). Unlike other forms of data collection discussed in this book, action research doesn't involve a particular strategy. Rather, action researchers may use many different strategies to gather data, both qualitative and quantitative, depending on the problem at hand and their personal preferences. In fact, action researchers are likely to use novel and eclectic approaches to gathering and interpreting data. Action research can be thought of as an approach to doing research, one with a very specific aim—the creation of social change—and one that involves creating particular kinds of relationships among all of those involved in the research process. In these respects, action research turns traditional ways of doing research on its head. Action researchers believe that the process of research must be open and democratic. Thus, they reject the clear separation between researcher and researched that the positivist tradition mandates. (Remember from Chapter 1?) Instead of viewing the researcher as the only one who can contribute to knowledge, they argue that ordinary people can as well. And instead of treating research participants as passive objects, like bugs under a microscope, they argue that research participants can become active in all phases of the research project, including formulating the problem and collecting and analyzing data. If individuals are the best sources of knowledge about their own lives, then they should become full and equal participants in the research process. Thus, participants are often called coresearchers, to reflect their more active role. Action researchers focus on both the process of conducting research and its outcome. Unlike traditional forms of research, in which researchers spec- 11V i" advance exactly what types ol questions they will ask (and the exact piocedures lor raising them), action research is typically fluid, allowing for i hanges as coresearchers together define what the problem is and how best in investigate it. The process of conducting research is liberating, as individuals gain knowledge about their own situations. Action researchers see knowledge as power. As individuals gain knowledge about the conditions al lecting their own lives, they gain the power to change them. The outcome of action research is also important. Action research typi-i ally begins with an interest in a specific problem of a community or group (Stringer 1996, p. 9). Action research thus tends to be local, specific, and oriented to a particular case (Small 1995). Its aim is always practical: to lead to some kind of change or practical application. For example, one potential outcome of the project that Lykes (1997) described was a photographic exhibit for the community. In this sense, action research is often applied research—research with real-world applications. VARIETIES OF ACTION RESEARCH An enormous variety of activities can be included under the umbrella term "action research." Action research is often multidisciplinary, bringing together people from diverse social locations and with diverse skills and abilities. While different people involved in action research may bring something different to the project, they all tend to agree that action—some kind of outcome—should come about as a result of their work. Research results shouldn't gather dust in the library, to be used only by people with advanced degrees. Research must be relevant to, and thus created by, those who are affected by it. At heart, all action researchers are concerned that research not simply contribute to our knowledge but also lead to positive changes in people's lives. Participatory Action Research (PAR) Within action research, there are several different traditions. One tradition, known as participatory action research (PAR), emphasizes the active participation of those being researched. One version of PAR was developed in Third World movements for human liberation, drawing on the adult education movement created by Paulo Freire (1970). This version of PAR is explicitly political; it emphasizes the political nature of knowledge production and sees the ultimate aim of research as human liberation. PAR practitioners in this tradition argue that knowledge production has typically been viewed as the province of elites, who use knowledge <<> control and oppress common people. One objective of PAR, then, is to create knowledge rooted in the lives and perspectives and experiences of ordinary people and directed toward progressive social change. Thus, PAR emphasizes investiga tion, education, and political action. The differences between this and other forms of action research are perhaps more of degree than of kind. PAR practitioners see themselves breaking down the walls between researcher and researched and creating genuinely equal participation. Perhaps more than other action researchers, their goals tend to be overtly political. So, for example, Noel Keough, Emman Carmona, and Linda Grandinetti (1995) describe a collaboration between Canadian and Filipino popular theater and environmental organizations, published in Convergence, a journal that focuses explicitly on PAIL Their goal was to improve the organizations' popular theater skills, to create public awareness of environmental issues, and to spur individuals to make changes in the size of their ecological "footprint." They worked collaboratively to disseminate information on sustainable development through theater performances and creation of a video and training manual. Other practitioners of PAR do not see themselves as having quite so overtly political a purpose. They emphasize the advancement of both basic-science and practical knowledge through the common participation of researchers and local communities (Small 1995; Whyte 1991, 1999). To make effective changes, they argue, some basic research is needed, and the results of action research conducted in this tradition are often published in academic journals and disseminated in other ways to reach a more general audience. These researchers often focus on organizations and may include decision makers in their research. For these reasons, the more radical PAR practitioners discussed earlier tend to distance themselves from them. Some action researchers in this tradition have worked with private industry and organizations (Small 1995). William Foote Whyte (1991) discusses several industry-based action research projects, including a project with Xerox Corporation and the Amalgamated Clothing and Textile Workers Union aimed at finding ways to cut costs without eliminating jobs or sacrificing workers' pay. With David Greenwood, Whyte was also involved in a study of organizational culture at the worker-owned Mondragon Cooperatives in Basque, Spain (Greenwood, Whyte, and Harkavy 1993; Whyte 1995, 1999). The Mondragon Cooperatives, which have made stoves, refrigerators, and washing machines since the 1950s, are one of the world's largest cooperative economic enterprises. Workers own the industrial plants, a coopera- iive bank, and other serviies unhiding research anil development and ed-iK.ilional services. Whyte and other researchers were interested in how w.ii kei-owned cooperatives could succeed in the long run, for conventional wr,.lom had argued that worker-owned cooperatives would inevitably fail oi remain small-scale. They began their PAR at the request of one of the . ooperatives, which asked for assistance in solving some organizational pioblems. The academic researchers were initially asked to teach the coop-eialive members research skills; the research eventually developed into a Luge-scale project on organizational culture (Greenwood, Whyte, and I larkavy 1993). IcministAction Research Feminist researchers also have a long history of doing action research. In fact, some people argue that feminist research is inherently action-oriented, because it is aimed at social change. Feminist action researchers date back to the early twentieth century, when the social sciences were in a state of rapid development at the University of Chicago (Reinharz 1992). Shulamit Rein-liarz describes several examples of feminist action research, including the work of sociologist Crystal Eastman, who, after receiving graduate degrees in sociology and law in 1907, studied industrial accidents with the goal of improving work conditions. Despite their long history of applied and activist research, relatively lew feminist researchers have thought of themselves specifically as action researchers (Cancian 1996). Although there are similarities between feminist research and action research, including an emphasis on political action and a critique of models of research that emphasize a hierarchical relationship between the researcher and those being researched, the two approaches have tended to stay distinct (Cancian 1996). More recently, however, some feminist researchers have begun to incorporate community participation as an important component of their research (Small 1995). Nancy A. Naples, for example, describes her experiences in a PAR project with childhood sexual assault survivors (Naples with Clark 1996). Herself a survivor of childhood sexual abuse, she had been active in a collective that tried to create alternative methods for healing. In her PAR, she wanted to fully involve other survivors and to merge her separate roles of political activist and researcher. J. K. Gibson-Graham (1994) also provides a recent example of feminist action research in her work with women in Australian mining towns. She recruited miners' wives to develop questionnaires and conduct interviews with other women in mining towns to collect stories about the women's activism in the Miners' Women's Auxiliaries ami olliei organizations. Evaluation Research Sometimes, researchers are invited to evaluate a program or policy, lor rv ample, drug counselors might want to see if their treatment program at tu ally helps addicts stop using drugs. Or a group of educators and school administrators might want to find out if their experimental school helps slu dents learn. Or social workers might want to see if their programs are serv ing those for whom they are designed. Or a legislator might want to know il an agency is providing services in the most efficient way possible. In eai h case, to do so, they need to conduct an evaluation. Sometimes, the practitioners may decide to conduct the research by themselves; other times, they may decide to bring in the services of an evaluation researcher. Evaluation researchers can use the same kinds of tools as other qualita tive researchers (see Patton 1990 for a classic textbook on qualitative evalu ation research; see also Posavac and Carey 1997]. They may decide to do interviews or field observations; they may analyze documents or do quantitative analyses. Often, evaluation researchers draw on the same collaborative techniques as other action researchers. They may, for example, work with practitioners to identify the focus and design of their research. When they work more collaboratively in this way, their work can be seen as having much in common with action research. RESEARCH FOR WHOM? Action researchers invite us to ask, Who owns the research process? In traditional research, specialized researchers, often connected with a university or educational institution, develop research agendas based on their own interests. They develop the problem statement, figure out what methods to use, conduct the research by themselves or with the help (often purchased) of others, and write up the results. The research participants are typically seen as "subjects," whose sole role is to provide the information the researcher is seeking. Although the research may be either basic (intended to contribute to general knowledge) or applied (intended to serve some practical application), those being researched don't typically have a voice in how the research is done or even how the results are to be used. In this traditional form of research, it might be said that the researcher "owns" the research process. A( lion research sets mil to demoi rali/.c the process. I low can a strategy l< >i i hange be effective il il isn't the creation ol those who are expected to !>6) calls a "social analysis" of the setting to ensure that all the necessary stakeholders are present. I ,et's say that you are interested in facilitating an action research project ■ m the parking problem at school. You know that there is a problem, be-. au.se you have heard many people complain about not being able to find a l '.ii king place, and you yourself have had that experience. The campus paper has published articles on the shortage as well. In this example, who might be the stakeholders? Certainly, anyone who has to park on campus might be a stakeholder. This includes commuter students most obviously, but it may include residential students as well. Faculty also park on campus, as do staff, which includes not only clerical and maintenance workers, cooks, and secu-i Ky guards but also administrators such as deans and provosts and chancellors. People who live nearby may also be stakeholders, as they find it difficult to park on the streets in their own neighborhood. Which of these groups would you need to invite to the table? Participatory action researchers might argue that you need to invite those who are disadvantaged by the current arrangements. Perhaps faculty and administrators have reserved parking spaces and don't have difficulty finding a spot. If that is the case, then you might want to focus on bringing to the table those who are more severely affected by the problem. (From this perspective, the goal would be to organize the disenfranchised.) Other action researchers might argue that you need to invite all to the table; otherwise, a mutually acceptable solution to the problem won't be found. L. David Brown (1993) describes this as the difference between northern and southern approaches. The southern tradition encourages participation by the poor and dispossessed and excludes participation by elites. The northern tradition may include a more diverse group of participants, including those who have organizational or societal power. In bringing together the stakeholders, you need to consider if there are any barriers to participation. For example, if you hold meetings at night, commuter students who have children or night jobs may not be able to attend. If you hold meetings during certain parts of the day, perhaps students will be in class and people in the neighborhood will be at work. Are there any language barriers? For example, the stakeholders may not all speak the same language. Are there any barriers of disability? For example, holding a meeting in a building without ramps or elevators will make it difficult for many disabled people to attend. What about social barriers? For example, if you are interested in working with a community group, holding the meeting on campus might be intimidating to members. You also need to consider whether any particular stakeholders are key (Stringer 1996). Some individuals may he especially inlluential in a parlk u lar setting (just like a gatekeeper in participant observation). It may be (hut, once one or two key individuals decide to participate, the rest of the coin munity will follow. Without those key individuals, you may not be able tu proceed. Formulating a Collective Problem Action research is by definition collaborative research. Thus, the research facilitator cannot single-handedly define the research problem. The definition of the problem must arise from the group. Randy Stoeker (1999) argues that community groups need to decide which decisions (such as how to define the question, design the research, and gather the data) to let the researcher make by him- or herself anil which to reserve for the group. This can entail a number of challenges. Most importantly, each stakeholder might have a very different understanding of the problem and potential solutions. To illustrate, let's return to our parking problem example. Students may see the problem as one of distribution. That is, there are plenty of spaces for staff and faculty, but not enough for students. Because they are paying tuition, they may feel that they have a right to a parking space. Administrators may see the problem as a financial one. They'd love to build more parking spaces, but there simply isn't room or money. Or they may see the problem as rooted in students' behavior, arguing that if students took the shuttle bus or if residential students didn't park on campus there wouldn't be a problem. Faculty may view their reserved parking spaces as an employee benefit. Finally, neighborhood groups may see the university as encroaching on their space and taking up valuable neighborhood resources without giving anything in return. To find out how the different groups see a given problem, the research group needs to gather data. Here, the research facilitator may play a key role in sharing information about specific ways to gather information, but the action group must decide how to gather the information and how to interpret it. At this stage, the group might use any one (or more) of the techniques discussed in this book, such as participant observation, interviews, or documentary research, as well as more novel forms. Sometimes, action research groups use quantitative methods as well, such as surveys. In our parking example, the group may decide to conduct interviews with students, faculty, and staff. It may also decide to observe in parking lots and to conduct documentary research—for example, tracing campus newspaper stories on parking or conducting an analysis of parking tickets issued. Think: Exploring (he IVolilciu In this phase, the research group works collaboratively to try to figure out what the data it has gathered mean. It's often a good idea to develop a de-.«i iplive account first, which defines how the various stakeholders see the process. At this stage, it may be helpful to convene a meeting (or a series of meetings) with groups of stakeholders, so that all have the opportunity to voice their opinions and contribute to the emerging analysis of the data. The ..search facilitator's role might be to help bring the stakeholders together and create a setting in which they can develop various interpretive accounts. Again, the stakeholders will have to determine which of the tasks they want to retain for themselves and which to leave for the researcher. In our parking example, perhaps the group convenes a meeting of students, faculty, staff, and community members who are all committed to working on the problem. They might break up into a number of small groups, each consisting of two students, two staff people, two faculty members, and two community representatives. Each small group might be given the task of defining the parking problem in a way that reflects all viewpoints. When the large group reconvenes, each group's work might be displayed. Then the large group might sift through the definitions, looking for commonalities and differences. In general, the process of sorting through the data will require several meetings so that all involved can have input. Groups may adapt any of the analytic strategies we discuss in Chapter 8 for this phase of the project. Act: Defining an Agenda for Action Finally, the time comes to craft a solution or an action step that speaks to the problem the group has defined. Clearly, some social problems are much more complex than others. Solving the problem of parking at a small college might be relatively easy compared with, say, solving the problem of access to health care or racism or environmental pollution. Still, at this stage, the group must try to define some action steps that will address the situation. They need to set priorities and figure out a plan of action. As with the other steps, this process should be done in a collaborative way. (This typically means more meetings!) The people who are affected by the problem must be the ones to craft the solution. The kinds of action steps depend entirely on the nature of the problem and the groups involved in the process. Some action steps are educational— for example, working to inform others about the problem. Often, action research groups devise novel ways ol educating I hose win > are allei ted, sin h as putting on plays, making videos, or writing comic hooks. for example, u* part of her work with Latina domestic workers, Pierrette I londagneu-Sotelo helped produce novelas (comic-strip-like booklets) that helped educate domestic workers about their rights (Hondagneu-Sotelo 1996). Some in tion steps are short-term; others may be long-term. What's most important is that the steps address the problem and implement the solutions artii u lated by those most affected by the problem. SHARING THE RESULTS In traditional research, researchers usually publish their results in an academic journal or book, and they may present them at a conference of academics as well. Typically, however, research results do not make it into the popular press (at least in the social sciences). Often, those who are the subjects of research do not find out what the researcher has said about them. This isn't because researchers don't want to share their results. Rather, it has to do with the structure of academic training and rewards. There aren't many rewards in academe for publishing in popular books or magazines or for presenting one's work on television or radio or via photo exhibitions in indigenous communities. Action researchers usually make a commitment to distribute the results of their work more widely. If they do publish in an academic journal, it is usually as an adjunct to other forms of distribution. Action researchers want to be sure that their work has an impact on the people who are most affected. SOME PROBLEMS AND OPPORTUNITIES IN ACTION RESEARCH For those social scientists who were attracted to the field because they want to effect social change, action research may seem to be a perfect solution. It allows them to use their specialized training in the service of social change. It brings with it the possibility of more equal and collaborative relationships than traditional ways of doing research. It allows researchers to develop community organizing skills even while developing research skills. Yet researchers typically aren't trained in those kinds of community organizing and educational skills. If you are interested in conducting participatory ii lion research, you may want to liml out more about community organiz-iii),', You might want to learn more about the 1 lighlander Center in Tennessee, which has long been involved in education for community change (( iaventa 1991; Greer 1991) or about Saul Alinsky's mode of community oiganizing (Alinsky 1971; McNeil 1995). Sharing Power Action research may also present other difficulties. One of those difficulties ■ enters around the issue of sharing power. To what extent can social researchers, who are often middle class, with highly specialized degrees, really share power with their collaborators, especially when they are from marginalized social groups (Chataway 1997; Lykes 1997)? And it's not clear to what extent marginalized groups are necessarily interested in such a sharing. As Cynthia Chataway (1997) notes, more powerful social groups often offer a role in decision making to less powerful groups as a way to co-opt or manipulate them. How can participants be sure that information won't be used against them? Establishing a coequal relationship can take extraordinary time and commitment, and even then it is not assured. On a practical level, difficulties can arise when a researcher analyzes a problem very differently from participants who are more directly affected by the problem or thinks that a very different solution might be needed. Should the researcher defer to the group? Or the group may express sentiments that seem contrary to the facilitator's values, as M. Brinton Lykes (1997) describes in her research in Guatemala. In one of her projects, she found that being accepted by the mostly male staff meant also being privy to language and jokes that were demeaning to women. She recounts, "I struggled with the unanticipated experience that in order to sustain my status as an 'insider' I was forced to violate my sense of self as a woman" (Lykes 1997, p. 734). In bringing together groups of stakeholders, a facilitator may have to deal with differences in power. For instance, in our parking example, how can the relative power held by faculty, students, neighborhood residents, and staff be dealt with? Students may see themselves as relatively powerless in relation to faculty. They may not feel free to speak openly in front of those who have the power to grade them. An action research session I was invited to had a similar problem. The goal of the two-day session was to find ways that the community could more effectively meet the needs of youths. Many small groups were established, each containing a few youths, a few community activists, and a few parents and educators. In a brainstorming session, the youths' voices were quashed as the adulls hetaine more and mote excited with their own vision. Leading Groups Action research is hard work. Because it is collaborative in nature, il in volves group leadership skills that are typically not taught as part of re search methods. Successful action research involves a sensitivity to group dynamics and an ability to work in a respectful way with diverse groups of people (McNicoll 1999). Thus, action researchers need to learn empa thy and listening skills. They need to figure out how to balance competing perspectives and to negotiate potential conflicts among the participants (Chataway 1997). Acting Ethically Action research also involves several unique ethical issues. As activists, action researchers are especially concerned with creating mutually rewarding, democratic relationships with their participants. Yet action researchers often stand to benefit from the work they do. They can gain status and earn promotions through their publications and grants. One of the issues that action researchers deal with, then, is the potential for exploitation of research subjects (Small 1995). Is it fair for researchers to reap career benefits from the research? Is research that does not seem to benefit those being researched ethical? In addition, because action research tends to be open-ended, changing in response to the needs and input of the community of researchers, it is difficult to know when (and how) to present action research proposals to institutional review boards (IRBs) (McNicoll 1999). If you make an IRB proposal before contacting community stakeholders, you will have trouble providing a general guide for what you will be studying, given that the problem definition will come from that very contact. But if you wait until after the problem has been defined, you may find yourself holding community members up while the proposal wends its way through IRB channels. As a rule, IRBs are not well set up to deal with open-ended research agendas. If you are planning an action research project at your school, you should consult with your professor or adviser to get additional information about how your IRB will treat such a proposal. Despite the difficulties involved, action research can be an exciting way to combine an interest in social research with a desire to effect social change. It can be heartening to see your research have a real impact—and to see yomsell allei ted by (he n-seau h plot ess as well. By working through die lesean h problem as a i oequal with others, by listening to their experi-• in es and developing a shared commitment to action, you can be trans-luf uied by action research. OUESTIONS FOR THOUGHT I Consider some of the differences between activist research and more traditional forms of researcher-directed research. Which form is more appealing to you? Why? What are some of the difficulties you might encounter in trying to do an action research project? What strategies can you think of to overcome those difficulties? d. What kind of a role do you think researchers should play in creating political change? Do you think creating political change should be a goal of research? Why or why not? EXERCISES 1. Consider a group that you might like to conduct activist research with. Outline a strategy that you might use to begin your research. How might you identify stakeholders? How might you get the process started while still allowing participation by the coresearchers? What strategies can you think of to ensure full participation? 2. Interview a community organizer. Find out what kinds of skills are needed for that work. What does she or he do on a daily basis? Then consider: How is that work different from and similar to action research? SUGGESTIONS FOR FURTHER READING For descriptions of participatory action research projects, see the journal Convergence, which is published by the International Council for Adult Education. Cancian, Francesca M. "Participatory Research and Alternative Strategies for Activist Sociology." In Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana and Chicago: University of Illinois Press, 1996. Patton, Michael Quirn,. Qualitative Evaluation and Rrsean Ii Methods, 2ml rd New bury Park, CA: Sage, 1990. Posavac, Emil J., and Raymond G. Carey. Program Evaluation: Methods mid ( aw Studies, 5th ed. Upper Saddle River, NJ: Prentice Hall, 1997. Reinharz, Shulamit. Feminist Methods in Social Research. New York: Oxford I ii.iv.i sity Press, f992. Focuses on feminist action research in Chapter 10. Small, Stephen A. "Action-Oriented Research: Models and Methods.". Journal o)'Mm riage and the Family 57:941-956. Stringer, Ernest T. Action Research: A Handbook for Practitioners. Thousand Oaks ( A Sage, 1996. 8 Making Sense of Data I he ultimate goal is to produce a coherent, focused analysis of some aspect of the social life that has been observed and recorded, an analysis that is comprehensible to readers who are not directly acquainted with the social world at issue. —Emerson, Fretz, and Shaw (1995, p. 142) I] - t Help! You've collected pages and pages of field notes and interview transcripts. You have gathered a number of documents. You seem to be swimming in a pool of paper. What do you do? How can you begin to make sense of this mass of materials you have collected? How can you begin to transform it into something coherent? How can you develop some kind of an analysis? In previous chapters, we focused on the process of collecting data. Now it's time to consider what to do with the data once you've gathered them. For many researchers, beginning data analysis can be difficult. The process of gathering data has its own rhythms, and it may seem hard to begin analyzing at the same time.] Still, ideally, you should begin data analysis in the field on in the process of gathering data. If you wait until you have finished the last interview or left the field completely, you won't be able to go back and ask clarifying questions. You won't be able to tailor your observations or your interviews as you make discoveries. Waiting until the papers pile up may also make the process of analysis more daunting. If you have been putting off trying to make sense of your data until everything has been collected—and in a qualitative research project "everything" can run to hundreds of pages— the process will be harder than necessary. Analyzing qualitative data generally involves several - stages.; First, you have to find some way to physically manage or organize the data: Put it into 151 three-ring binders or liles, enter it into the computet, 01 develop some othei technique. Then you need to immerse yourself in the data and become familiar with what you have gathered. As you become increasingly intimate with the data, you will begin to generateahemes or categories Or identily patterns in the data. Finally, you will need to find ways to present the analv sis to others. MAKING MEANING OUT OF QUALITATIVE DATA In qualitative research, data analysis is a process of making meaning. It is a creative process, not a mechanical one (Denzin 1989]. It's not as if all tin-meanings are embedded in your field notes or interview transcripts or documents, and your job, as researcher, i^to "uncover" them. Rather, your job is to actively create meaning out of your" raw materials. Two researchers, encountering the same data, may analyze them very differently. This is to be expected, given that you bring to the process different skills, emphases, and theoretical orientations. At the same time, the different ways of analyzing the data should at least be plausible. Consider the process of creating something out of a set of raw materials. For example, given a bolt of fabric and some notions (thread, zippers, buttons, and so forth), different people might end up making very different kinds of creations. One person might make a dress, another a Halloween costume, and a third a tent. But what they come up with must at least be plausibly created from the raw materials. It would be awfully difficult to make a working motorcycle or an edible meal out of a bolt of fabric. Here's another example: My office is a terrible mess right now. There are books and journals scattered all over—in piles on the floor; on bookshelves, and so on. There are several filing cabinets of different colors. On one of the two desks are a few coffee cups and a plate. There are pens, pencils, reams of paper, and a box of Kleenex. Running clothes hang on a hook on a wall, and a shirt is draped over one of the chairs. There is a computer, a printer, and lots of standard office supplies, like rulers, scissors, tape, and computer disks. Suppose I ask you to organize my office. You might do so in any number of ways. You might lump all the books and journals together as "things to read." You might place the desks and chairs in the category of furniture. You might place the pens, pencils, and computer together as things to write with. But clearly this is only one way to organize my mess. You could also categorize things by color: all the beige things (computer, printer, one of the hhng cabinets), all the blai k things (tape dispenser, calculator, the other filing lahincl, some books), all the colorful printed things (running shorts, lilale, Kleenex box, some books), and so on. Or you could divide them into two i ategories: things to throw out and things to keep. How you decide to oiganizc the objects depends very much on your purpose in categorizing them. If you are planning a yard sale, arranging them by color won't do you much good. You'll want to know what should be kept and what should be gotten rid of. But if you are trying to make sense of how I use this office and i lie things in it, arranging the things by function (things to write with, things to read, things to use for running) might be helpful. Or, if you want to distinguish among things that I use often and things that I don't use, you might want to arrange things spatially. Items on the tops of piles or near my computer are things that I use often. Books on shelves are less often used than books on the floor. Pencils left on top of the desk are favored over those that are in a drawer. Again, your purpose in analyzing the materials will suggest a strategy. One of the difficulties in analyzing qualitative data is that there is no single method for making sense of what you have found. Just as there is no single "right" way to organize my messy office, so there is no single "right" way to organize and analyze qualitative data. MANAGING DATA As you begin the process of analyzing qualitative data, your first task is to arrange or organize your data so that you can begin to make sense of it. Margaret LeCompte and Jean Schensul (1999b) call this process Jtidying up." That is, you have to do some "serious housekeeping" before you can begin the analysis. You need to make sure that all of your taped interviews are transcribed and that your field notes are together and complete. You need to label the computer disks or the audiotapes so that you know what is contained on each. You need to make a comprehensive list of all the materials you have gathered. At one level, this is a fairly mechanical process of gathering all your materials and devising a filing system or other way of organizing them so that you can easily access them. Yet even at this level you can go in various directions. Separating Different Types of Data You may want to keep different types of data separate. For example, you may choose to separate transcripts of interviews from your field notes. I usually assign each interview a code number as 1 conduit the interview Then, as interviews are transcribed, I put a copy of each transcript in a 11 It -folder along with a copy of the face sheet and any Held notes relating to the interview. I also store any correspondence or other materials relating, in that interviewee in that file. And I usually keep these files separate from my field notes, which are organized chronologically, and other documents. Keeping Data in Chronological Order Some materials, like field notes, should be kept in chronological order. I tisn ally type up full field notes on my computer and print out copies during the course of my observations. In a three-ring binder, I place all my field notes chronologically. I also sometimes put into this binder copies of any documents I have collected during the course of my research. (It's best to do this while you are in the field; on the day you collect a document, write the date on it and put it into the binder.) If I am collecting a series of documents, such as a set of newsletters, I'll usually keep these in a separate file, organized chronologically or, for undated materials, alphabetically. If you are keeping a personal journal for the research, you may want to place copies of the dated entries in the same binder as your field notes, again organized chronologically. Some people like to keep these separate; the choice is yours. As you are generating these materials, be sure to label them clearly. For example, write "FIELD NOTES" or "PERSONAL JOURNAL" at the top of the page. I often keep original documents (including handwritten scratch notes) in a separate file. You don't need to use a binder (or multiple binders for a big research project) if you don't find it helpful. A file box will work, too. Whatever system you use, make sure you devise a system for keeping the papers together and in order. Organizing by Topic or Document Type You may want to organize some of your materials according to a specific topic. For example, if you have collected a number of written documents, you can organize them by topic rather than by chronological order. Suppose you are studying a mothers' organization. During the course of your field research, you collect a number of miscellaneous documents. You might have a number of handouts and magazine or newspaper articles about different topics: methods of discipline, arts and crafts activities, and stories about mothering. You could organize these according to topic. Or you might have several types of materials: meeting agendas, newsletters published by the gioup, newspaper i lippings, and llyeis adv.arising the group. You could ar-i.mge these according to type ol document. IM^king a List or Logbook I lowever you decide to arrange your materials, it's a good idea to make a comprehensive list of everything you have gathered. For interviews, you might want to include the dates that the interviews were conducted and tianscribed, as shown here: Interview 001 002 003 Date Conducted 2/15/99 2/15/99 2/19/99 Date Transcribed 3/1/99 3/4/99 3/5/99 For documents, you might want to include the title of the document, its date, and a short description of it, as shown here: List ofT)ocuments ) 1. Working Mother Forum (photocopied newsletter) April 1999 May 1999 July 1999 August 1999 September 1999 October 1999 January 1999 February 2000 March 2000 April 2000 May 2000 June 2000 2. Meeting agendas and minutes April 1999 through April 2000. December 1999 missing notes and agenda because meeting was canceled. 3. Miscellaneous documents a. Mission statement and bylaws of group b. Handout on birth order given out at April 1999 meeting c. Mother's Limenl poem e-mailed lo me I»v "No/" May IM'l'l d. Handout on hiding day care given mil al luiir I1)!)1) meeting Some researchers like to keep track of their materials in a computer daia base; others like to use index cards. You should use whatever system is nn f.i comfortable for you. Choosing Between Computer and Hard Copy If you use a software program for qualitative data analysis, you need lo enter your data in a form that the program can use (we'll discuss qua I it a tive analysis programs later in this chapter). Many qualitative researches find that a basic word processing program works fine. Even if you are going to be doing most of your analysis directly on the computer, it's still a good idea to print out a hard copy (and often two or three copies] of all your materials. Keeping Multiple Copies Make sure you keep multiple backup copies of all your materials. Hard drives can fail, computer viruses can erase files, and accidents can happen, I usually make several backup disks of the interview transcripts and keep one at my home office and the other at my school office. I know one researcher who had just returned from the field and hadn't yet had a chance to make copies of much of the material. During a routine fire drill, he carried all of his data outside with him "just in case." Most researchers have heard horror stories (whether true or not) about researchers who lost all of their data to a catastrophic fire or a stolen car or a flooded office. The moral of the story? Make multiple copies of your field notes, interview tapes, transcripts, and other data, and keep them in different places. You should keep one copy as a "master" file and two or three other copies for different kinds of analyses (Patton 1990). Some people—those who are organized—have an easier time with this stage of the research process than others. If you are good at keeping things filed and in order, this stage will be much more pleasant than if you tend to be disorganized and hate to put things away.* Still, the process of gathering your materials together and organizing them is important. You can't analyze *If you are a compulsive organizer, you may enjoy reading Werner and Schoepfle's 1987 book Systematic Fieldwork, Volume 2. If you are considerably less organized, you may want to stay away from it! ,|.ii;i thai v*'ti can'l hid. Whalevei method you choose, be sure lliat you ,iv,cmble all of your data and can tnnl what you need, when you need it. And be sure to keep multiple copies ol everything. (ill TING INTIMATE WITH YOUR DATA The next stage involves getting intimate with your data, Maybe some time has elapsed between when you gathered your data and when you're ready lo begin the analysis. Even if you have been keeping up, some time has probably passed since you began the research project, and interviews and observations from the beginning of the project may have faded from your memory. The goal at this stage is to immerse yourself in the data. Read over your field notes from beginning to end—several times. Do the same with interview transcripts and any documents you have collected. If you learn better by hearing than by reading, you might find it helpful to listen to the interview tapes—several times if necessary. Some people like to listen to their interview tapes in the car. As you review your materials, take notes if you find it helpful. But don't worry at this stage if it feels as if you will never be able to make sense of all of your materials. Your main object is to try to load up your memory with all your data. You should think about it while you're driving and while you're eating and while you're running and whenever you have a spare moment. CODING Once you begin to immerse yourself in the data, how do you make sense of them? How do you identify what is most important? How do you figure out what to leave alone? How can you develop a sense of which themes are recurrent? How can you reduce the sheer volume of material to a more manageable quantity? The first step in making sense of your data is coding. You might already have some ideas about coding based on quantitative methods. In quantitative methods, the goal is to come up with a small number of categories so that you can manipulate the data on the computer. You begin with a set of categories and assign numbers to cases. For example, if you asked interviewees how much education they had, you might get a variety of responses worded in various ways. One person might say that she had a college education; another might say that he graduated from high school; another might say that she had an associate's degree In quantitative uuling, you would assign each one a number—lor example, a " I" lor less than a high mIkiuI education, a "2" for a high school diploma, a "3" for some college education, and so forth. The process is mainly a procedure for counting how many pen pie had each level of education (Coffey and Atkinson 1996). In qualitative analysis, the goal is not to assign numbers to cases. Ratlin, the goal is to begin to focus on the potential meanings of your data. Amanda c°ffey„and Paul^dmson suggest that qualitative coding entails three hush "procedures: "(a) noticing relevant phenomena, (b) collecting examples of ^ those phenomena, and (c) analyzing those phenomena in order to find com monalities, differences, patterns, and structures" (1996, p. 29)^ Especially at die 'Beginning of the process, you'don't want to limit potential insights by rigidly applying preestablished codes to your data. Instead, you want to use the process of coding to begin to reveal potentiul meanings. My students often find that initial bouts of coding add more complexity rather than simplify things. They sometimes despair that they'll never figure out what their data mean. This is to be expected. What you want to do at this first stage is to begin to develop hunches and ideas about what is going on in your data. Later, as you refine your codes and develop basic themes, the meanings will become clearer. Open Coding Many researchers use some version of grounded theory to work with their data and develop meanings (Strauss and Corbin 1990). Essentially, this method involves a two-stage process of coding. In the initial stage, called open coding, you work intensively with your data, line by line, identifying themes and categories that seem of interest. In this early stage, you should JSBaiil OEgn to whateverjou see jnjie^ata. Thus, you shouldn't hesitate to note categories or themes that may not seem relevant to your original research problem. In open coding, you don't use someone else's preestablished codes, or even your own. Rather, your goal is to see what is going on in your data. If you develop codes in advance, you will impose your own sense of what ought to be there in the data and may very well miss what is there. In developing open codes, it's important to make sure that you understand the particular themes and categories you see in the data. As you work through your data intensively, don't worry if you see somewhat different things and use different labels for your themes. As you become more familiar with your data, you will naturally begin to see patterns and commonalities and develop a focus. """ " ... - You can do open coding in a number of ways, for example, you can simply write your cock's in the margins. II you like working directly on the computer, you can use the comment function of a word processing program to insert codes. Some people like to use multicolored highlighters or pens to note key phrases in conjunction with marginal notes; others like to use col-oied I'ost-it notes. You should try a few different strategies to see what works best for you. figure 8.1 shows an example of open coding based on a short segment ■ if an interview conducted with a mother who was juggling paid work and household work. In this segment, she talks about what that juggling was like lor her. A number of themes seem prominent: finding day care providers, assessing the quality of day care providers, leaving her son, and so forth. Development of Themes 3 "^'J After you have done open coding for a while, some recurring themes should begin to emerge. Maybe some categories show up in case after case (in interviews) or over an extended period (in field notes). Maybe some themes seem especially interesting or relevant. At this time, you need to work more intensively with your codes to see what kinds of themes you might develop. Out of all the various codes you have created, which ones might be most helpful in shaping your analysis? How will you know when to do this? There's no single "right" time, and you can repeat this process as often as you like. It's usually helpful to have done open coding on more than one case (for example, if you are coding interview transcripts, it's helpful to have gone through several different interviews) or a number of days' worth of field notes. When you begin seeing the same codes over and over and don't seem to be seeing quite as many new themes or creating many new codes, this should suggest potential themes. There are any number of "tricks" you might use to develop themes. Which ones you use will depend in part on what kind of learner you are. Some people like to see things visually; others like to physically manipulate things. Some people work best on a computer; others do best on the floor surrounded by their materials. You should probably try several different approaches until you figure out what works best for you. Some analysts like to physically sort through the categories they are developing. As they are doing their open coding, they create a series of index cards. Each index card contains the relevant line or portion from the interview or field note on one side and the code on the other. They make sure to note exactly where each piece of the data is from (for example, the Ml' Mi Mi K: Can you talk a lictle about what it was like for you when you were both working, and with your son, when you were trying to balance that. What was that like? R: First of all, finding day care was very very difficult for me. I interviewed a dozen people. Um. Most of them were youngs young ladies with new babies themselves. And it was very stressful to think about leaving my son, whom I wanted to have a child, leaving him to be raised with somebody else. I ended up finding a day care person who was in her mid 50s. She was really Ml jtt>'"t> ■JH 1,1"' Mi' > »1* a nice woman and I felt very comfortable with her. And the whole < FIGURE 8.1 Open Coding interview number and page number). Then, when they have done a significant amount of open coding and have a number of cards, they sort them into piles to see which codes emerge most frequently. They might sort them a number of different ways to see if the same kinds of themes seem to emerge. Each time they sort the cards, they keep track of which categories they created. (There are also computer programs for creating and sorting index cards that can be very helpful.) Instead ol making index i auk, .'.nine researchers literally cut up a copy <'l the interview transcript or their held notes into pieces, each with a rele-\ nil line or segment marked with a code. Again, they carefully note where ■ .it It slip comes from so that they can refer back to it. Then, they sort the hps into piles, often multiple times, to see what themes seem to emerge. And, again, they keep track of the outcome of each sorting. focused Coding j t nice you begin to identify several key recurring themes, you can do focused i oiling (Coffey andAtkinson 1996). Like open coding, focused coding entails going through your data line by line, but this time you focus on those key themes you identified during open coding. So, for example, in the open coding segment in Figure 8.1, I might decide to focus on two themes: day care providers and separations. I would then go through a copy of each of the transcripts and field notes, line by line, noting those categories. Of course, if you were doing open coding on a longer section of data (as you should), you would probably develop more and perhaps different themes. As with open coding, you can use any technique you like for focused coding. Again, some people like to write directly on the text; others like to work on the computer. There are a number of software programs for qualitative analysis that can help you store and retrieve codes; you may also choose to use an ordinary word processing program. If you are working with word processing software, it's often helpful to do a computerized search for key words or phrases. You may need to check your searches carefully, however. Sometimes, people use somewhat different words to describe similar things, and unless you search for all the possibilities, you may miss important sections of the text. For example, if I were looking for passages relating to day care providers, I would probably need to search a number of different words: "day care provider," "child care provider," "baby-sitter," the name of the specific provider mentioned by the respondent, and so on. If it's available, some people like to work with a clean copy of the data in this stage. I sometimes find it helpful to create a long word processing document with all the quotes I've identified for each code, as in this example (with all my transcripts and notes already on the computer it's a fairly easy process): Day Care Providers—Feelings About "She was a really nice woman and I felt very comfortable with her." (001, p. 4) "The- day care woman was really nice, she very laielv < ailed me loi any trivial thing." (006, p. 4) "I was so angry with [my day care provider]. When my daughter got bit by her cat, I was out of there. I couldn't believe that she wouldn't just lock the cats up." (009, p. 6) "I really like how [my day care provider] is so calm and patient with Zoe. She's terrific with her. I always lose patience." (010, p. 15] Thus, if one of my codes is "feelings about day care providers," I might ere ate a word processing document that contains each quote that relates to the code. (I'd be sure to indicate the interview number and page number so I could find it easily.) I might then break that down into further categories: positive feelings, neutral feelings, and negative feelings, or perhaps strong feelings and weak feelings. When I was ready to write up the analysis, I would already have the quotes in one place. Other researchers might make index cards or use other techniques. You are probably getting the impression that coding is a laborious, time-consuming process. It is. But in qualitative analysis, there really are no shortcuts in analyzing your data. While the computer programs available for data analysis can help you manage data and reduce the amount of time you spend retyping and rewriting things (such as making index cards), they can't do the analysis for you. You still have to develop the codes and themes yourself. You still have to make the decisions about what is important and what is not. At the same time, there are many different techniques you can use; there is no one single way to analyze qualitative data. This leaves you free to develop the methods that make the most sense for your particular situation. Maps and Diagrams Many researchers like to use visual techniques in working with their data. For example, you might look at the themes you have developed and create a map or diagram that shows how the themes seem to relate to one another. There are any number of ways you can create diagrams or maps. You might, for example, map how individuals or groups relate to one another. This means identifying the various groups in a social setting and determining whether members of one group are also members of another. For example, can you be both a student and an employee at the same time? Do some individuals serve to "connect" groups or bring them together? You may want to diagram the structure of events in a sequence of action. For example, consider children's birthday parties. What are the vari- Chldien aitive af parfy \ give pre*enfa fa Cpuf a way W; eh.? ein o&tivif) 7 (.trap*, &f>en ploy? Uttivify tarlt rely m oil kid* taming- of tame- time? \ 'ml' vme4, eh. tiriair fill oil children pre*enr? . game*- limbv, mwfital ohir*, pin He fail 'Plifl&S *9metim0* giver\ fa winner* i f?efjre*/Menf*< alway* include* take and usually i*e treat* \ Mare ploy/atfivifie* Lsptwnal? \ Tre*enf* Cn$f opened of ene parfy, seemed ®dd? \ 6ive g&odie bog* \ Children g& hme FIGURE 8.2 Map of Events at a Children's Birthday Party ous events in a birthday party? What are the things that must be present in order for one to consider it a birthday party? There are usually games, things to eat, birthday cake, "goodie" bags, and presents, among other things. But would it still be a birthday party without a cake? What about a party without the guest of honor—the child who is having the birthday? Is there a specific order in which events happen? Or are there several possible or-derings of events? Figure 8.2 shows a map of events at a typical children's - patents &I birfUay Mj — bitrHday t>k(d relative* e>Her adulh - family friend* — relative*? C^randparents, aunts/undes, eto.? - parent"*? guests entertainers — tlffWnS, musician, ^ee pa'inter FIGURE 8.3 Map of the Relationships Among Attendees at Birthday Party birthday party; Figure 8.3 shows a map of the relationships among party attendees. Creating a diagram of the flow of events can help you figure out if you have included all the steps in an event or process and if you understand the process fully. MEMOS At each step of the way it's helpful to write up memos. Writing is a process °^ ^--^™5a,ning- Writing memos helps you shape your thoughts about the emerging analysis and provides a record of your progress. Thus, through the process of writing about your findings, you develop your analysis a bit further. Think of memos as letters or notes to yourself to help you understand your data. They are written for you, not for a teacher or an outside i < ;u 1<-i.' I In is, you lau use wh.ili'vri loimat you like ami leel most eomlorl-able with. Whatever loimat you use, however, it's important to date your memos so you know when you i reated them. Procedural Memos You can use memos in a number of ways. First, you can write up procedural memos to help you remember how you did your coding, what kinds of categories you created, and so forth. (Strauss and Corbin [1990] call these code memos.) Procedural memos focus on the nuts and bolts of your research. They summarize exactly what you did—what codes and categories you created, what they included, what codes you rejected and why, and so on. These memos are important for they help you keep track of what you have done. (They are also useful for creating a description of how you analyzed your data, which you will need to include in your write-up.) Some people like to keep a separate logbook of their memos, which helps them keep a running record of all the steps they took. Others prefer to keep their memos with their field notes. I find it helpful to give each memo a title so that I can keep track of it. The following memo distinguishes between two codes: Procedural Memo: Feelings About Day Care Providers and Pick-ups and Drop-offs The code "feelings about day care providers" includes any kind of emotion word or evaluation that interviewees use to refer to their day care providers. These include things like the provider was "nice" or that the respondent "liked" her, and so forth. It applies to feelings about the provider as a person (or as a provider). It does not include feelings the interviewee might have about placing her child in day care or feelings about pick-ups and drop-offs. The code "pick-ups and drop-offs" includes anything the respondent says about picking up or dropping off her child at day care. Ex: "I was always glad to let my partner do the drop-off" or "I liked to go in a few minutes early to see how my daughter was playing with the other kids." Analytic Memos ) You can also write analytic memos that help you think about the categories and themes you are developing. In these memos, you focus on what's important in your data and make i onnections between eases. II \">u'n following the advice I gave in Chapter 4, you've already been writing lhl». kind of memo while doing held work. These memos contain your hum In n and ideas and best guesses about what you should be thinking about Ah you develop your coding further, these memos should get more and mm, detailed. Ideally, you should be able to draw on these memos in your lin.il write-up. Here's an example of an analytic memo: Analytic Memo: Drop-offs It seems like the drop-offs are very charged for some of the mothers I've been interviewing. Some of the parents seem to have a hard time with it—I think they feel guilty about leaving their kids in day care (maybe they feel like bad mothers?) and the drop-off then becomes difficult. This was definitely the case with Roz and Lee ("I can't stand it when she cries 'Mommy stay, Mommy stay' Every morning I wonder if I'm doing the right thing"). Not all the mothers find it hard, though. Leah said: "I love to see Zoe [her daughter] throw herself in Callie's [the day care provider] arms. She really loves it there." Does it have to do with the quality of the day care provider? The parents' own feelings about their mothering? The parents' feelings about their work? Maybe mothers who enjoy their work more or need the money more feel less ambivalent about the drop-off. [I'll need to check this out in field notes too. Do I notice the same kinds of difficulties?! ASKING QUESTIONS Coding is not the end of your analysis, however. It's more like a beginning. In the initial stages, the coding opens you up to the possibilities of your data; it makes it more complex. Coding thus serves the purpose of what Amanda Coffey and Paul Atkinson (1996) call "data complication." At the same time, focused coding helps you to reduce the data—to bring it down to size and make it more manageable. Yet this data reduction, by itself, is not the finished analysis. You still have to answer what I think of as the "so what" question: So what? What do these data mean? You still have to figure out what the themes and categories that you are identifying might mean in conjunction with one another. At this stage, as you are working intensively with your data, it's helpful to ask a number of questions (Emerson, Fretz, and Shaw 1995; Patton 1990). And you can focus your questions in a number of directions: truest inns about evenlv What happened.' Who was involved.-' I low did the event begin? I low did it end? (Questions about chronology: What happened first? Next? Then? (Questions about the setting: What is this place like? What does it look like? Questions about people: Who are the people involved? What are they like? Questions about processes: What are people trying to accomplish? How do they do this? Questions about issues: What are the key issues for these people? What is important to them? How do they describe what is important? What language do they use? Answering these kinds of questions can alert you to different facets of the analysis and help you describe an event or a group of people. For example, if you have been observing at mothers' play groups, you might want to describe what happens at a typical group meeting. This kind of descriptive writing is important, as we'll discuss further in Chapter 10, for it helps you to set the scene. Part of a good qualitative analysis entails describing to others what you have observed. Answering these questions can also lead you to make interpretations about the meanings of things. For example, you may want to interpret what the play groups mean to the women involved in organizing them. Perhaps they serve as a way for the mothers to provide friends and playmates for their children or as a source of support for the women. Perhaps they help solidify the women's identities as "good" mothers. These are all interpretations—arguments about what the play groups mean. These kinds of interpretations must be grounded in the data—hence the laborious process of working intensively, line by line, with the data. Some scholars argue (hat till writing is iuleipielive That is, even win n you are describing something, you can only present youi interpretation ul what you saw. Regardless of where you come down on this issue, you should always think about how you can orient yourself in terms of two modes ul inquiry: (1) how to describe events or people or cases to others and (2) how to interpret (or make your best argument about] what those things might mean. DEVELOPING AN ANALYSIS Developing a qualitative analysis may seem like a mystical process. As I keep emphasizing, the activities you engage in throughout data analysis are not mechanical ones. But there are several methods of analyzing data that may be of use to you. These activities generally include looking for patterns in the data (similarities and differences), comparing cases, building typologies, and conducting a content analysis. In Chapter 9, we will consider the pne cess of constructing narrative accounts or stories. Looking for Patterns One of the first things you can do in your coding is to look for patterns in the data. Different interviewees might deal with the same kinds of issues or handle different issues in similar ways. The same kinds of events might occur over and over, or similar patterns might occur among different events. The coding that you do should help alert you to these kinds of patterns. For example, in the analysis of working mothers, a researcher might find that women give similar accounts of leaving their children with the day care provider. Perhaps all of the women describe a process of dropping their children off in the morning and picking them up at the end of the day. Comparing Cases Once you have identified a few patterns, you may want to compare cases more systematically. What is a case? It all depends on your research. If you have primarily conducted interviews, you might want to compare individual interviewees. Each person you interviewed might be a case. You might also compare events, like birthday parties. Each birthday party might be a case. Or you might consider cases on a larger scale: organizations (like universities or colleges), neighborhoods, and so forth. I'll k 111 • 1 Irop-off • .i-ll .Iocs 1)1)1, 0(1.', ()()/, 1)08, 001, 002, 005, 006, OK', 013, 014 008, 011, 015 1 '.u i ncr does 003,005,006,011 003, 007, 012, 013, 014 1 -li i partner 004, 009, 010 004, 009, 010 (interviewee does both) 1 K ;t I RH 8.4 Chart for Comparing Cases For example, let's say you have interviewed fifteen mothers who work outside the home. You may want to systematically compare them on a variety of dimensions: whether they tell pick-up/drop-off stories, whether their spouse or partner shares the work of picking up and dropping off, and so lorth. You could create a grid or chart like the one in Figure 8.4 to help you compare. Here, you're simply comparing the cases in terms of who picks up and drops off children at day care. For example, your first interviewee (001) does both the pick-up and drop-off. Interviewee 3 does neither—her partner does both. Interviewee 7 does the pick-ups and her partner does the drop-offs. The single parents in the sample (004, 009, 010) naturally do both because they don't have a partner to share the work with. Comparing the cases in this way may help you think about other ways that you can compare the cases. For example, some of the women who have partners are like the single parents—they do all the work of picking up and dropping off. Are they similar in other ways as well? Different? Building Typologies One relatively simple analysis involves the construction of a typology, which is simply a system for categorizing types of things. So, in our day care example, you might notice while coding that some of the women seem to share picking up and dropping off their kids at daycare with a partner and others do not. Maybe you notice that some of the women seem to share the work of child care more generally with their partners and others do not. You might want to create a typology: equal sharers and unequal sharers. If you look at the grid in Figure 8.4, you can see that in three cases the woman does both pick-up and drop-off (001, 002, 008) and in one of the cases her partner does both (003). You can call this group the unequal sharers. In seven of the eases, the woman ami her partner split op the picking tip mill dropping off equally (005, ()()(>, 007, 01 I, 012, 013, ON). These are lit" equal sharers. In three of the cases, the woman doesn't have a partnei, »u you might make them either a third category (single parents) or leave theni out of the typology. Obviously, this is a very simple typology. Sharing pick-up and drop oil might not say much about whether the women's partners share in othci activities, such as staying home with a sick child, supervising homewoik, feeding and bathing the children, and putting them to bed. But it might help you think about ways to make sense of the data. If some partners share thin task equally, do they share other tasks as well? You might go back through your interview transcripts and field notes to find out. And how do people feel about the tasks that they share—or, perhaps more tellingly, don't share?* Getting Stuck and Unstuck It can be very easy to feel "stuck" while you are attempting to analyze the data. Maybe you have been coding for a while and haven't "come up for air" to look at the general patterns. Then, when you do, it's hard to get ahold ol them. Maybe you're having trouble telling others what you think you art-finding. If you do feel stuck, here are a few tricks that may help you become "unstuck." First, you might try making analogies of the form "this phenomenon I am studying is like X, and unlike Y." For our birthday party example, you might say, "Children's birthday parties are like adult birthday parties in that the person whose birthday it is usually gets presents and friends are invited to come celebrate. Children's birthdays are unlike adult birthdays in that children are often happy to get a year older; but adults often are not." If you are really stuck, you might try coming up with analogies that seem ridiculous. Second, you might try choosing different words for your key concepts or codes (Lofland and Lofland 1995). Sometimes, the way things are phrased obscures meaning. For example, if you have been studying peer pressure, try saying "pressuring peers," "influence," "force," "fads," "crazes," "role modeling," or "following the crowd." Each way of wording gives you a slightly different spin on the concept. Third, try talking about your developing analysis with friends and fellow students. They can often help you clarify what you are seeing or stimulate new ways of thinking about your topic. Listening to others talk about how "If you're interested in "equal sharers," you might enjoy Francine Deutsch's 1999 interview study of parents who share the work equally, called Halving It All. diey aie dealing with a puiliiulai tlala analysis problem might give you ■i>me ideas as well. It's olten lu-lplul to think about how you might explain your developing analysis to someone who knows nothing about your topic. I ••■ «im-limes suggest that students think about how they would describe their ■I udy lo a 12-year-old child, an older relative, or a being who has just arrived 11 < hii outer space and knows nothing about their topic or the culture. CONTENT ANALYSIS < >ne technique that can be used to analyze any kind of text is content analysis, which generally involves a systematic analysis of texts. Texts include any kind of written material, such as books, magazines, diaries, letters, minutes ol meetings, transcripts of TV programs, interview transcripts, and field notes. Often in content analysis, quantitative scholars count the frequency with which specific words or themes appear in the texts and then conduct statistical analyses on those frequencies. Qualitative scholars may use content analysis as well. Sometimes, simple counting is useful. For example, if you want to know how often men and women appear in textbooks, you might simply count how many photos there are of each. This strategy is sometimes used by people who want to document discrimination against particular groups or by those who want to see if stereotyped images are improving over time (see, for example, Clark, Lennon, and Morris 1991). You may want to use this kind of simple counting in your field notes or interview transcripts. For example, it might be helpful to count how often a particular theme is mentioned by your interviewees. Suppose you want to gain more understanding of pick-ups and drop-offs at day care. You might count how often each interviewee mentions a pick-up or drop-off and, for each case, decide whether the event is negative, positive, or emotionally neutral. You can then get a sense of whether, overall, the interviewee sees each event as positive or negative. At other times, you may want to focus more on the meanings of texts. For example, Ralph LaRossa and his colleagues wanted to understand how fathers have been portrayed in the media and whether the cultural meanings of fatherhood have changed over time (LaRossa, Jaret, Gadgil, and Wynn 2000). Thus, they decided to conduct a content analysis of Father's Day and Mother's Day cartoon strips published between 1940 and 1999. They found that the cultural meanings of fatherhood had changed over time, but not in a straightforward way. Depictions of nurturing fathers, for example, were more likely to appear in the late 1940s and early 1950s, as well as in the 1990s. What should you count? It all depends on your research topic. (See Carley 1993 for a discussion of some of the choices you must make in i on tent analysis.) Some content analysts count specilic words (Ryan and Wei* ner 1996). Others might count every time a theme—for example, leellng positive about one's work—appears in a phrase or a sentence or, more rarely, a paragraph (Berg 2001). Still others count individuals—for example, whether an article is authored by a male or female or whether photos In elude Asian Americans or African Americans or European Americans. Still others might count concepts, things that can't be observed directly, such its "guilt" or "deviance." Researchers who use content analysis may focus on several different levels. At the simplest level, researchers stick to relatively straightforward content—words that appear in the text directly or pictures or objects that can be counted directly. This is sometimes referred to as manifest content. Others try to conduct a more interpretive analysis, focusing more on the underlying meanings in the text. This is known as latent content. Still others focus on the linguistic or semantic properties of the text itself (Franzosl 1990a, 1990b). They argue that the way in which things are expressed also gives insight into the meanings of texts. SEMIOTIC ANALYSIS Researchers sometimes use semiotic analysis, the study of signs (Manning and Cullum-Swan 1998). A sign is simply something that stands for something else. So, for example, a rainbow flag might symbolize the lesbian/gay/ bisexual movement, the ® symbol peace, and the Star of David (# ) Judaism. A sign consists of two parts: (1) the sign itself (such as the flag or the star or the peace sign) and (2) the content with which it is associated. Semiotics is based on language, but there are other semiotic systems as well. Thus, mathematics is a semiotic system, as are highway signs and systems of etiquette, which might include things like shaking hands or bowing to greet someone (Manning and Cullum-Swan 1998, p. 252). By themselves, signs don't mean anything. They only take on meaning within a particular social context and when interpreted by someone within that context. Signs thus can take on many different meanings and may be connected in a number of different semiotic systems. In semiotics, the analyst tries to "unpack" the meanings of the signs. These meanings may be multiple and layered. You "read" signs all the time even though you may not be aware of it. For example, I dropped my daughter off at kindergarten this morning. All I lie i hildrcn were wearing uuilonns: white blouses and shirts, blue plaid -.kills and navy slacks. The youngest children were in matching navy blue sweat suits with the school logo printed on the front. The school logo itself is a sign that symbolizes the school. The uniform symbolizes order and uniformity, as well as perhaps a desire to level class differences in the hoof Uniforms may be associated with parochial schools or with public « barter schools that emphasize the qualities of discipline and order of a parochial school. In a public charter school, then, the uniform might be a siy;n of a sign. In addition, all the children in my daughter's school wore backpacks or carried bags with logos indicating particular brand names (signs of social class) or characters like Winnie the Pooh, Tigger, Barbie, and the Powerpuff {iirls. Each of these characters can also be analyzed: Winnie the Pooh stands lor childhood and innocence, Barbie can be analyzed in terms of race and class, the Powerpuff Girls symbolize girl power, and so on. Parochial schools themselves can take on many different meanings. They may signify religious faith and the desire to inculcate certain religious values in children. They may signify educational quality. In neighborhoods with inferior public schools, parochial schools may signify a neighborhood-based alternative. They may also signify class distinctions between those who can afford private schools and those who cannot. There are many different ways to conduct a semiotic analysis and a number of different theoretical traditions that shape the kind of analysis you might do. Readings in semiotics are often very difficult. Many who do this kind of work have been trained in literary criticism or cultural studies. If you are interested in finding out more about semiotics, you may want to see the article by Peter Manning and Betsy Cullum-Swan listed at the end of this chapter for a relatively accessible introduction. GROUNDING THE ANALYSIS Let's say you've gone through several stages of coding and writing memos and analyzing. You have worked to try to see how the various codes you developed might relate to one another. You have drawn maps and diagrams and feel you have developed what seems to be a sound analysis of your data. Perhaps you have conducted a semiotic analysis as well. What next? How do you verify that your interpretation of the data is actually a good one? Before you finish your analysis, you will need to examine your work carefully against the data you have gathered. You want to make sure that the analysis you've developed is actually supported by the data. Shortcomings in the Research Michael Huberman and Matthew Miles list a number ol "shorU omings" In qualitative research (1998, pp. 198—199). First, researchers may gal her Inn much data, or too much of a certain kind of data, so that they fail to see put terns or "skew" the analysis in one direction or another. You need to considei carefully whether you've collected the right kinds of data or whether you'i r missing crucial information. Second, first impressions tend to be more intense than later ones. You may weigh initial impressions more heavily than you should. The same li true for more recent impressions. Often, you tend to remember the most recent events, not the overall pattern. In conducting your analysis, you need to pay attention to events taken from the whole time span. After you have completed an initial analysis, you need to check to make sure you have done so. Third, people tend to see things that occur at the same time as related when it may simply be coincidence. For example, suppose that every morning when you get up you see your neighbor leave in her car, presumably to go to work. Does your getting up have anything to do with your neighbor? Of course not. Sometimes, qualitative researchers see related patterns when none, in fact, exist. Fourth, the information may not be reliable. Perhaps an interviewee didn't tell the truth or exaggerated. Perhaps you didn't take good field notes and don't have a good record of what you saw. Perhaps you didn't interview the right people. You need to think about what kinds of shortcomings you have in your data. Some of these shortcomings are fixable. If you are still in the field, you can gather more data. If you have field notes spanning a long period, you can try to focus especially carefully on the middle events. Sometimes, simply knowing that people have a tendency to make unrelated events seem related can cause you to scrutinize your interpretations more carefully. Sometimes, however, shortcomings in the data or the analysis aren't fixable. If you've already left the field and cannot go back, you cannot patch the holes in your field notes or interview more informants. In these cases, it is better to be honest in your write-up about the problems you have discovered. Negative Cases One of the first things you can do to ground your analysis more carefully in the data is look for negative cases. In creating your argument, you probably amassed positive examples (evidence) that support your case. For example, il I believe that women's identities as mothers is implicated in the difficulty "i ease with which they drop their ihildivn oil at day care, I need to look l«a positive cases in which mothers express both difficulty with drop-off and guilt about their mothering role. Having found those cases, however, I il.< i need to look for evidence that disconfirms my interpretation. For example, ii 1 find a series of cases in which the mothers express great difficulty with drop-offs but also express satisfaction with their mothering role in gen-i ial, I probably need to think again about the analysis I created. If you are .till collecting data while you are conducting your analysis, you can specifically seek out negative cases. Regardless, you should always search through your data for negative cases after you complete the initial analysis. The Null Hypothesis Trick I loward Becker (1998) suggests that you should initially assume that there is no pattern or relationship in the data. Then, you need to tease out the conditions under which the pattern you observed might develop. He calls this the "null hypothesis trick" because in quantitative research researchers are always trying to support the null hypothesis, the hypothesis that two events or variables are not related. In a qualitative analysis, you might want to do something similar by assuming initially that there is no pattern in the data. You then have to amass the evidence, based on examples, to show that there is a pattern. So, for example, the null hypothesis in our child care research might be that there's no relationship between dropping off one's child at day care and how one feels about oneself as a mother. In order to feel reasonably secure in the interpretation that there is such a relationship, you would need to come up with clear evidence, based on examples from the data, that there is one. In doing so, we would need to sort through the selection processes that lead people to end up in various social groups. (Becker's book, Tricks of the Trade, offers many useful examples of how you can work with your data.) Here's a simple example. Let's say you want to understand the process of how some parents come to be full-time parents after the birth of a child while others return to the workplace relatively quickly. The null hypothesis assumes that there really aren't any differences between the two groups. Looking over your data carefully, however, you notice that men are more likely than women to return to the workplace quickly, though many women do as well. You might focus your attention, then, on the question of why? You might notice that women are more likely to have access to paid maternity leaves than men, who have very little access to paid parenting leaves. You might notice that men often get paid more than women, so that it costs a family more to lose the man's income. You might notice thai some women and men enjoy their jobs very much, ami others don't. You might also |m.|. at the kinds of social messages about parenting and work that men and women receive. By paying attention to the social processes that treat men and women as very different sorts of parents, you can begin to umleisiaud some of the differences you observed between the two groups. Triangulation Triangulation is often used to mean bringing different kinds of evidem eh. bear on a problem (Denzin 1989). Thus, if you have access to interview data, observational data, and historical documents, your analysis is likely to be much sounder than if you rely on only one source of evidence. This is because each kind of evidence has its own strengths and weaknesses. With observation, you can actually see how people behave; it allows you to see a whole process unfold over time. With interviews, you can gain insight into their feelings or reasons for behaving in a certain way. Using multiple kinds of data allows you to balance the strengths and weaknesses of each. So, loi example, if you want to know about the social messages that male and female parents receive, you might want to analyze the media. Yet this kind of analysis will tell you relatively little about how individual women and men feel about those messages or whether they take them into account when making decisions about their own lives. Triangulation—using different types of evidence—will help. Some researchers also use the term to refer to the findings of different researchers (Huberman and Miles 1998). Thus, if you come to a similar conclusion as other researchers, you can feel more confident of your findings. II you come to a very different conclusion, it doesn't mean that your analysis is wrong. It may mean that you used different kinds of evidence, perhaps from different historical periods. Maybe you emphasized very different themes or were informed by very different theoretical traditions. You may want to examine the methods others used in conducting their research to give you insight into any differences you find. COMPUTER-ASSISTED DATA ANALYSIS Just about all qualitative researchers use computers for at least some aspect of their field research. While there still may be a few who use typewriters to write up field notes and transcribe interview tapes (or who write up all their notes by hand), these ate suiely a dislui. I (and dwindling) minority. Most i. se.iiclieis at least use woid pi messing sol I ware to transcribe interview i.ipes ami write up lull held notes. Many will use simple search capabilities . .1 sorting procedures. In that sense, computers are used widely throughout • liialilative research. Yet relatively few qualitative researchers use software ih.it is specifically designed for qualitative data analysis (Richards and Ivu hards 1998). But that situation may be slowly changing, as a growing number of computer programs have been developed especially to assist in qualitative data analysis. Why haven't qualitative researchers been faster to incorporate the . omputer into their procedures for analyzing data? Probably most impor-i antly, the computer can't solve easily the kinds of problems that qualitative lesearchers confront. Qualitative researchers work with texts (words), not numbers, and computers can't "crunch" texts as easily as they can numbers. Qualitative analysts rely on their own judgment, which can't be wholly mechanized. Thus, you already have to know what you want to do with vour data before using a qualitative data analysis program. Dedicated programs for qualitative data analysis can also be expensive and difficult to learn. For these reasons, I often recommend that beginning researchers work with their texts by hand or on word processors before attempting to use a qualitative data analysis program. That way, you can develop a feel for what it is that you want automated and whether a data analysis program might help. That said, computers (and qualitative data analysis programs) can be extremely useful. They can do the kinds of mechanical storage and retrieval tasks that we often do by hand. Computers have far better memories than humans do, as well as greater capacities for storing texts. Used well, computers can offset some of the shortcomings involved in qualitative data analysis, such as relying more heavily on later field notes than on earlier ones (Dohan and Sanchez-Jankowski 1998). Qualitative data analysis typically involves (1) recognizing patterns (or categories) in the data, (2) generating ideas about what these patterns might mean, and (3) exploring potential meanings in the data. Computer software programs can help in this process in several ways (see Dohan and Sanchez-Jankowski 1998; and Richards and Richards 1998 for useful overviews). There are numerous software programs available (Dohan and Sanchez-Jankowski count over twenty). None, however, is dominant, and the availability of programs changes rapidly. To figure out which program you might want to use, you need to identify your purpose and try out a variety of programs. ■ lunges i. nine i.ipidly, liy (lie 11111 > - 11 * 1 Unit, is | iiihlisliei I, any general review ill sullware programs will he mil ol dale. Whether y?■■ 2. Have you spell-checked the paper? 3. Are. your citations in appropriate form? Are all paraphrases and direct quotations appropriately credited? Have you double-checked to make sure that all necessary references are included? 4. If you included appendixes, have you mentioned them in the body of the paper? AUDIENCES AND VENUES Writing is not only a process of discovering (or developing) what you think but also a means of communication. Although some writing is personal, intended only for your own use, a qualitative research report usually will be read by others. After all, you have done all that work of collecting and analyzing qualitative data. You now have to think about how to share your work with others. In doing so, you need to think about who your audience or your intended readers are. There are many dill'erenl audienees you might consider. ( M'ten, iicmlrm ics are a major audience for qualitative research. A great (.leal ol qualitative research is written with them in mind. But practitioners (such as so, lul workers or educators) and the general public are also potential audiences. For example, if you have been studying workers in a homeless shelter, the shelter staff may be very interested in what you found. Perhaps you have been studying personal networks among international students (Sia 2000) If so, the international student office at your school might be interested In the results. Perhaps you have been conducting an evaluation of a program 01 doing participatory action research with a group of activists. If you have re^ ceived sponsorship from a particular group or organization, they will prohn-bly expect a report. In all of these cases, you need to consider how to frame your results in ways that are useful to those audiences. If you are writing for other academics, there are a number of venues In which you may present your work. Academics present their research at professional conferences (like the annual meetings of the American Sociological Association and the Midwest Sociological Society), in scholarly journals, and in books. As a rule (though there are exceptions), academic writing tends to be more formal than writing for a general audience. In academic writing, you usually need to include information about the methods and theories you used. In writing aimed at a general audience, you generally write more informally and include less information about your methodology. (You might notice, for example, that some books based on ethnographic or interview data include a section on methods in the appendix; that way, the more academically inclined reader can find out more and the general reader can ignore it.) If you are writing for practitioners or a business audience, you may find that you need to write more concisely. Practitioners tend to be very busy people, without a lot of time to waste. Thus, your main findings should be condensed. This chapter focuses primarily on writing for an academic audience. If you are writing a paper for a class, it may help to imagine an audience that consists of other beginning social science researchers, like yourself. Assume that your audience doesn't know very much about your particular topic but does know a little bit about the process of conducting social research. Thus, your audience will want information about the specific methods and procedures you used but will not need to be told, for example, specifically what a focus group or an interview or a participant observation is. You should also assume that your reader is a bit skeptical. Your reader doesn't believe anything on faith, but instead wants to see enough of your data and methodology to come to her own conclusions about your findings. USEFUL WAYS TO STRUCTURE QUALITATIVE RESEARCH Qualitative research reports traditionally have been more flexible than reports of quantitative research. This is, in part, a product of necessity. Because qualitative researchers cannot fall back on tables or numbers or charts to present their data, their words have to convey the analysis. Thus, qualitative researchers have to think more carefully about how they say things than quantitative researchers do. They have to pay attention to language. In recent years qualitative researchers have begun to present their work in more creative formats (Richardson 2000). Some qualitative researchers have begun to experiment with fiction and poetry (for example, Denzin 2000; Krieger 1991). Others have experimented with dialogue and personal narratives (Denzin 1998; Ellis 1998; Ellis and Bochner 2000; Ronai 1992). These innovative ways of presenting research are often depicted as an alternative to the traditional research report. And they are. But it still helps to have a sense of what a traditional research report looks like. A traditional research report typically has these parts: title (don't forget to make a title page), abstract, introduction, review of the literature, data and methods section, findings, and conclusion. References are usually included in a separate list of works cited. Sometimes, appendixes will present additional information at the end of the paper. Title and Abstract The title should give a good sense of what the paper is about. Ideally, it should be catchy as well, inviting the reader to pick up the report and jump in. You might want to scan the titles in the references of this book. Which ones seem to give you a good idea of what the article or book is about? Which ones are more obscure? The abstract is a brief description (less than 250 words) of the main points of the paper. It should include a summary of the most important things you found, the evidence you cite, and an indication of how you analyzed the data. Essentially, an abstract should give a potential reader just enough information to decide whether to read the paper. In very traditional academic writing, aulhois ate lomiseled not to ns. personal pronouns (like "1" or "we") in abstracts (or, indeed, throughout th. body of the paper). This has led to a tendency to write abstracts in llie ,.„., sive voice ("It was found that . . ."). The use of the passive- voice lends n, imply that no one actually did the research; it somehow merely appeared m "was found." But I strongly recommend writing in the active voice. It's bet ter to use "we" or "I" to show that a real person did the research. The following is an abstract of an article that appeared in the join mil Gender & Society. Although a little unwieldy, the title describes what the article is going to talk about: "Gender and Emotion in the Advocacy lot Breast Cancer Informed Consent Legislation." The abstract is short- less than 150 words—and gives a summary of the full article. This is a qualitative study of the role of gender and emotion in a political setting. The data are from interviews of activists and legislators, as well as from archival accounts of the debates in state legislatures about breast cancer informed consent legislation. I found that proponents for and against the legislation shared the belief that women are more emotional than men. This social belief shaped the political strategies the activists adopted and initially contributed to their effectiveness; however, their opponents claimed that the women activists should be dismissed because their emotionalism made them irrational. I close by discussing Western cultural beliefs about emotions, social stereotypes regarding women, and their consequences for women activists in political arenas. (Montini 1996, p. 9) Notice how the author includes a brief description of the approach she has taken to her topic (the focus on the role of emotions and gender). She describes her data sources (interviews and archival research) and sketches out her main finding: that beliefs about women's greater emotionalism both helped and hindered the movement for legislation. Introduction Just as its name implies, the introduction should introduce readers to your topic. It should give them a hint about what is to follow and entice them to continue. You might want to think of an introduction as a slippery banana peel. You want readers to "slip" on your introduction and "fall into" the next section. In qualitative research, you have a number of options for your introduction. Beginning writers are often counseled to begin with a general statement and move to the specific. In the introduction, this might take the form il documenting why tlx- topn ol youi papei is ol general interest and then noving on to the spei ilii s ol yoin study You can use this kind ol standard ipem-r for a qualitative reseau h report, but you have other, more creative iplions available as well, for example, you might begin with a specific case a a story about a research participant that illustrates the main points you will be discussing. You might begin with a story about how you entered the held or encountered the problem. Or you might begin by setting the scene: describing your field setting. Notice how Joshua Gamson begins his study of tabloid television talk shows: Let's begin here: talk shows are bad for you, so bad you could catch a cold. Turn them off, a women's magazine suggested in 1995, and turn on Mother Teresa, since watching her 'caring feelings' radiate from the screen, according to psychologist Dr. David McClelland of Harvard, has been shown to raise the level of an antibody that fights colds. (1998, p. 3) Notice how this beginning pulls us in. We want to know whether Gamson really means this. It also sets up a discussion of his aims, which he outlines in his opening chapter. Josepha Schiffrnan uses a slightly different technique in her study of grassroots peace organizations. She begins with a scene from her field site: It's a rainy San Francisco night, and an affinity group of Bay Area Peace Test (BAPT) is meeting to plan an upcoming act of civil disobedience. One woman volunteers to facilitate, and the group collectively constructs an agenda. First they deliberate alternative actions, making sure that everyone has a chance to express an opinion. They also discuss how they're going to maintain a sense of community in the Nevada desert, how they will support each other through the various hardships they're likely to encounter: radioactive dust, dehydration, possible police brutality. The meeting ends with an evaluation of the group's process, and finally, holding hands, they sing a freedom song. (1991, p. 58) This introduction gives us a sense of the group. Schiffrnan then moves on to a description of the second organization she is studying, followed by a more theoretical description of her aims in the article. Compare these introductions to Theresa Montini's introduction to her article on breast cancer informed consent legislation: This article is a qualitative study of the role of gender and emotion in a political setting, taking as a case example women's activism for breast cancer iiilormcd consent laws. I studied loiniei bieasl laneei patients who worked lor the passage ol breast earner inloimed eon sent legislation in their state legislatures during the decade ol the 1980s. A breast cancer informed consent law specifies that alter a woman is diagnosed with cancer, but before she is treated lor cancel, her practitioner will give her information regarding the various treatments that are available and appropriate to treat her cancer and gain her specific consent for the administration of the treatment. The procedure may sound like standard protocol in any practitioner-patient encounter, but it is not. Public concern prompted legislation to be introduced in 22 states and passed in 15 during the 1980s. (1996, p. 9) Which introduction is best? It depends on a number of factors, including individual preferences in writing style, the audience, and the kinds of data that follow. If you have been conducting participant observation, for exam pie, a personal account of how you first encountered the group may be more fitting. If you have been conducting a textual analysis, that kind of per sonal story may not feel appropriate. If .-.i.lve the seeming contradiction. I'erhaps will solve some of the pressing problems seen i loti inhadii I ol In -i s, 11 y In i a new theory or new lei hliiqiie identified in the literature. Sometimes, qualitative researchers use the literature review section to outline a theoretical perspective. Because qualitative research typically toe uses on a small number of cases or on a unique field setting, there may not be a large body of research on exactly the same topic to draw on. A researcher might want to explore how a particular theoretical perspective might help shed light on her or his case. For example, Theresa Montini (1996) was interested in using a sociology-of-emotions perspective to understand a particular social movement: the movement to gain informed consent laws for breast cancer patients. The literature review should help your readers understand the context for your research. A badly written literature review can be the most boring section of the report. No one wants to read through pages and pages of sentences that all sound something like "Author X found Y; then author Z found Q. Author A found B, and author C found D." Here's a hint: If you are bored writing this section, then your readers will be bored reading it. Literature Review At some point in your report, you have to link your work to the broader research on your topic. In more traditionally organized papers, the literature review is included as a separate section, usually near the beginning. In this section, you outline the main controversies in the literature. You note the main questions other researchers have asked, the theories and methodologies they used, and the results they obtained. Your goal is to show how your work fits into the larger body of research. You need to think about how your work relates to the main controversies and debates in the field. Does it challenge existing ways of thinking about the topic? Does it extend previous studies? Does it try out a new methodology? Does it use a new field setting or source of data? When well done, a literature review section leads readers to think that your work is a natural and logical step in research about the topic. I strongly recommend that you organize your literature review themati-cally—that is, according to the main themes or debates within the literature. For example, you might open with "There are three main controversies in this field: A, B, and C." You can then discuss how each of the studies you've found relates to A, B, and C. You may want to be critical. For example, if some of the research is badly done, explain the shortcomings. If some studies Sources of Data and Methodology The next section of the paper typically deals with the question of methods. This section should include a description of the field setting, if you did an ethnographic study, or whatever other data source you used. Methodolo-gists sometimes say that this section should provide enough information that someone else, armed with your report, could replicate your research. Of course, in qualitative research, this clearly isn't possible. If you observed in a field setting during a particular time and place, that time has passed already. A new researcher entering the field will encounter a changed setting, even if the only apparent change was the entrance (and exit) of the former researcher. In addition, observational research depends on the personal relationships between the researcher and those being researched. In this sense, qualitative research cannot be replicated. You cannot duplicate the exact relationships and events that occurred. Still, you should give enough information in your methods section that another researcher could try to approximate what you accomplished. In general, you need to discuss the particular methodological choices you made and the reasons you made them. If there were particular issues related to confidentiality or protection of the research participants, you should describe these as well. Sources of Data first, you need lo discuss die soun <■ ol your data. I! yiai conducted interviews, you should specily how you located interviewee',, how many interviews you conducted, with whom, and under what (ii< urn stances. You should also specify where the interviews were conducted, how long they lasted, how structured they were, and whether they were taped and transcribed. When it's relevant, you should also include demography information about the interviewees: age, sex, race, and so forth. If you analyzed documents or other material artifacts, you should stall' what kinds of documents or artifacts you used and address issues of data collection. How did you find these artifacts for analysis? Are there any pai ticular problems with the artifacts? For example, are parts missing? Ate there problems of authenticity? If you are using a series of magazines m newspapers or other published materials, you should include information on the issues selected. Why did you select the particular issues you did? Are there missing issues? Why? If you conducted observational research, you should include a description of your setting and how you gained access to it. You should specify for how long a period you observed and during what specific times. You should include basic information about what you did during the period of observation. For example, if you observed in a public place, did you openly take notes? Methods of Analysis You should also describe the procedures you used for analyzing the data. Did you conduct a content analysis, or analyze your interviews for the main themes, or use a grounded theory approach? In comparison to quantitative researchers, qualitative researchers tend to give much less information about how they analyzed their data. Still, you should at least mention any specific procedures you used. If you used a software program for qualitative data analysis, you should mention the name of the program and any related details that seem helpful. Again, keep in mind that your main goal is to help other researchers understand what you did. How much detail is enough? How much is too much? You might find it helpful to imagine potential readers for your research. In general, you can assume that your readers are much like yourself. They have a basic understanding of qualitative methods, just as you do. They don't need to be told, then, exactly what participant observation is or what an interview is. You can assume that they know. But because there are different kinds of interviews, they need to know what kind you used. They don't need to be told what observation is; they need to know how you accomplished it within a particular context. I lie Main llody: landings The main body ol your repoii should locus on your findings. This section should he the heart of your paper It's what you've spent most of your time dying lo figure out, alter all. The main body can be structured in a variety of ways, depending on your research. Sometimes, researchers organize this section according to the main themes they found in their analysis. Let's say, for example, that you lound two distinct styles of leadership in your observation of a student organization. You might want to organize your findings around the two different styles you found. Or suppose that, in your analysis of the image of teenagers in the news media, you found four main ways that teenagers are portrayed. You might want to organize your findings around those four ways. Another way you might organize your findings is as a narrative, especially if you have been conducting participant observation. In this case, your main goal is to tell an ethnographic story. You might choose, for example, to tell the story of one of your research participants, or you might tell the story of a specific event. The story serves to highlight something important you discovered in the field. If you choose this strategy, ask yourself: What is the most important story I want to tell? How can I tell it in a way that will be believable? If you organize your results as a narrative, you may find yourself weaving your analysis of it into the story. (Remember, the analysis is the "so what" part; it tells why this story is important.) Or you might decide to place your analysis in a separate section, following the story itself. However you organize this section, remember to include support for your analysis. This support should be in the form of examples from your research, including quotations from your interviews or field notes and excerpts from your documents. How many examples is enough? How many are too much? There are no magic numbers. You need to include enough examples so that your analysis is believable. But you don't want to include so many that readers will get bored or lose the thread of your story. As you're writing this section, you need to think about levels of generality. At the broadest level, you can make universal statements, statements that are meant to be about "human nature." As a rule, most qualitative researchers hesitate to make these kinds of claims—and for good reason. We simply don't have the kind of data (or maybe the chutzpah) to make such claims. At the other end of the spectrum, you can make very specific statements about a particular incident or a particular person. In qualitative writing, you typically have to move back and forth between different levels of generality. You can use specific statements about incidents or persons to try to back up broader theoretical claims. And you can use the insights from the data you collected to reflect on (or at least generate ideas about) laigel groups of people. Because the main body is usually the longest section ol your report, yon may want to create headings to divide the different parts within it. I lending help readers figure out where your argument is going. They're like street signs: although you may be able to figure out where you are without them, it's much easier if you have them. Summary and Conclusion The summary and conclusion should, just as the name implies, briefly summarize your most important points. You don't need to discuss everything you found, just the most important points. In this section, you might also discuss the implications of your research. Are there any policy implications? Is further research indicated? What next steps might researchers want to take? List of References and Appendixes At the end of your report, you should include a list of references, providing publication information for all works cited in the text. Different academic disciplines have different styles for referencing others' work. Most journals and publishers have specific guidelines for the format of citations. In general, sociologists tend to use parenthetical style rather than footnotes. That is, they include the author and the year (and page number, if they are quoting directly or paraphrasing] within the body of the text and provide the full citation in a separate section at the end of the report. I strongly recommend that you consult a writing handbook or style guide to make sure you use the correct format. Sociology students may want to obtain a copy of the Style Guide of the American Sociological Association, which is available from the American Sociological Association (www.asanet.org]. What should you reference in the text? In general, you need to provide citations to acknowledge the work of other authors. If you use someone else's written ideas in your own report without acknowledging the source, that is plagiarism—whether you intended to plagiarize or not. You should always include a citation when you make a direct quotation from someone else's work. You should also include a citation when you paraphrase another author (that is, restate their idea in your own words]. Consider these examples: Direct quote: "Careful citation is important because plagiarism is burglary—cheating which presents another writer's words or ideas as if they were your own" (Sociology Writing Group 1991, p. 45). Paraphrase: According (o (he Sot iolugy Writing t iroup ( 1991, p. 45), plagiarism can be consideied a form ol burglary or theft because it involves passing someone else's words off as your own. To avoid plagiarism, writers should be careful in their citations. When in doubt, it's a good idea to err on the side of inclusion. That is, it's better to provide a citation that you may not really need than to omit one that you do. After the list of references, you may want to provide various appendixes to your report. These may include, for example, a map of your field site, a list of questions or topics covered in interviews, photographs from a magazine or other text you have been analyzing, or any other supplemental information that might be of use to your readers. You don't want to include too much, however. For example, you probably shouldn't provide complete transcripts of all your interviews, although you might want to include portions. (If you are doing a narrative analysis, you should probably include portions of the transcripts.) MAKING WRITING VIVID The best qualitative writing is vivid. It tells a gripping story; it provides an insightful analysis; it tells us something vital. The best qualitative writing is like a good story, a page turner you have to stay up all night to finish. Why, then, as Laurel Richardson (2000) asks, is so much qualitative writing so boring? Partly, she answers, because we have not "put ourselves in our own texts" (p. 925). We have not developed our own voices but have, instead, relied on formulaic and scientized ways of writing. One of the challenges of qualitative writing is trying to tell the story in your own words. How can you find an authentic voice of your own in which to write? This is a tricky proposition, of course, for you may have been taught over the years not to do this. Thus, many students (and professional sociologists as well) avoid using "I" or "we." They use too much jargon and inaccessible language, thinking that this will make them sound more "educated." They "puff up" their writing with extraneous words. Many use the passive voice (for example, "It was found that . . .") in order to make their work seem more "objective." These are all things you should avoid. The only way you can develop your own voice is to write a lot. (Reading a lot helps, too.) Practically speaking, you may find that keeping a journal is one way to get that kind of practice. You should also try to avoid jargon. If you can use a short word, why use a longer one? If the longer word is more precise, then hy all means use it. But il'tlie short word serves just u* well, use it. It will make your writing leaner and help readers move through it faster. My guess is, as well, that you are more likely to speak and think In shorter words; they may be closer to your own personal voice. It may nlto help to refer to yourself in the first person ("I" or, if appropriate, "we") It sounds awkward and a little arrogant to refer to yourself as "this author." Another way to make your report more compelling is to tell the story in your research participants' own words. Often, a direct quotation or stoiv will make the point better than you could otherwise. Susan Krieger (1!)H3), for example, chose to structure her book-length exploration of identity inn lesbian community using only the voices of her research participants Except for the first chapter, her entire book consists of carefully selected quotations and paraphrases from the interviews she conducted. Although this is an unusual strategy (most researchers speak in their own voices us well), it highlights the possibilities for creating an analysis out of the participants' own words. Whether you attempt a more experimental or more traditional text, try to weave the voices of your participants into your writing, See if you can make your research participants visible throughout the text, OTHER GENRES OF QUALITATIVE RESEARCH Certainly, there are other ways to write qualitative research. As I have mentioned at various points, some qualitative scholars have experimented with fiction, poetry, dialogue, and a variety of other styles. Critics of more conventional techniques argue that traditional scientific writing is dry and dull and that it takes the person out of the research (Ellis and Bochner 2000). These critics argue that the new techniques for presenting qualitative research, like autoethnography, poetry, and fiction, are a way to "merge art and science" (Ellis and Bochner 2000, p. 761). Some critics, like Norman Denzin (2000), maintain that the boundaries between fiction and nonfiction, ethnography and story, and poetry and performance art are artificial—they are all social and political constructions (p. 899). Qualitative scholars still debate the place of fiction, poetry, and autobiography in qualitative research (see, for example, Gans 1999; Karp 1999; Plummer 1999; Whyte 1996). In general, those who are more comfortable with a postmodern perspective have been more open to and willing to embrace fictional techniques. From their perspective, there is no single "right" story, but merely a number of different stories that might be told. Fiction provides a useful adjunct for these scholars. For others, especially those who are working within a positivist tradition, the use of fiction vio- lates the tenet that in studying the empliii al world researchers should be objective and ground their studies in the "real" world. You will have to figure out for yourself where you stand in this debate. Whatever decisions you make, remember that writing is a process of discovery. Writing helps you figure out what you think. It is also a method for communicating what you think to others. QUESTIONS FOR THOUGHT 1. Think about your own writing routines. Where and when do you do your best writing? What kinds of routines do you have for beginning writing? How can you try to establish good work habits for your writing? 2. What kind of advice have you been given about writing? Have you been told, for example, to "write from your experience"? Or have you been encouraged to use "big words" to sound more "academic"? How might the advice you have been given aid you in or discourage you from developing your own voice? EXERCISES 1. Write an introductory paragraph to a qualitative study. Then try rewriting it using a different technique. For example, if you used a standard opening ("This paper is about . . .") for the first version, try using a description of an important event or key research participant for the second version. Which version do you like better? Why? 2. Write a paragraph that describes what your research is about. First, try writing it without using the personal pronoun "I."Then rewrite the paragraph in your own voice. Compare the two versions. Which version do you like better? Why? 3. For one of the paragraphs you wrote in Exercises 1 and 2, count the number of words you used. Then take a pencil and see how many words you can delete without changing your meaning. Again, count how many words you used. What is the smallest number of words you can get by with? 4. Find a piece of writing that you think is interesting and well written. It can be fiction or nonfiction. What makes it well written? Can you emulate some of those qualities in your own writing? 5. This exercise assumes that, you have been assigned to write a pupei lm your class, lake a first draft ol your paper and switch copies with a i las^ mate. Act as an editor for your classmate, and have your classmate do tin same for you. After you have read the paper, answer these questions a. What is the author's main argument(s)? Your goal here is not to criticize but to restate the author's points. (This will help youi classmate evaluate whether his or her argument is effective.) b. What evidence does the author use to support his or her argument? (Be specific.) c. Does the author's main argument seem well considered and c red ible? Why or why not? How might the argument be improved.' d. What are the greatest strengths of the paper? e. What are the greatest weaknesses of the paper? f. What suggestions do you have for improvement? 6. After you have received your classmate's feedback on your paper, make a revision plan. Try to answer these questions: a. After you completed the first draft of your paper, what did you feel you had accomplished well? What did you feel needed more work? b. Overall, what did the feedback indicate? What were the major strengths and weaknesses? c. What are the most important changes you need to make for the final draft? d. How are you going to accomplish those changes? 7. Take an excerpt from your paper that presents an analysis or description. (This should be part of the findings or main body.) Rewrite the excerpt in the form of a dialogue between two people. Then, rewrite it in the form of a poem or fictionalized story. What are some of the differences among the three forms? Which do you like better? Why? I ..iniott, Anne Hin/ /'v lUul Suimi' histitti /ie//\ mi Writing ami life. New York: Aik Inn, I 90S. A general (ami vei v well vvi it leu) sell help hook on writing. Kuliardson, Laurel. "Wiitiiig: A Method of Inquiry." Pp. 923-948 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage, 2000. Includes a number of useful suggestions for making your writing more interesting and provides some exercises to help you experiment with more novel methods of data presentation. Sociology Writing Group. A Guide to Writing Sociology Papers. New York: St. Martin's Press, 1991. A good, basic guide for undergraduate students. Wolcott, Harry F. Writing Up Qualitative Research. Newbury Park, CA: Sage, 1990. A short, chatty monograph that provides practical advice for writing up an ethnographic project. SUGGESTIONS FOR FURTHER READING American Sociological Association. Style Guide, 2nd ed. Washington, DC: American Sociological Association, 1997. Becker, Howard. Writing for Social Scientists: How to Start and Finish Your Thesis, Book, or Article. Chicago: University of Chicago Press, 1986. One of the standard guides—and an excellent model for writing. Janesick, Valerie J. "Stretching" Exercises for Qualitative Researchers. Thousand Oaks, CA: Sage, 1998. Contains many exercises that help you focus on writing as a method of analysis. Appendix A American Sociological Association Code of Ethics Ihe American Sociological Association's (ASA's) Code of Ethics sets forth the principles and ethical standards that underlie sociologists' professional responsibilities and conduct. These principles and standards should be used as guidelines when examining everyday professional activities. They constitute normative statements for sociologists and provide guidance on issues that sociologists may encounter in their professional work. ASA's Code of Ethics consists of an Introduction, a Preamble, five General Principles, and specific Ethical Standards. This Code is also accompanied by the Rules and Procedures of the ASA Committee on Professional Ethics which describe the procedures for filing, investigating, and resolving complaints of unethical conduct. The Preamble and General Principles of the Code are aspirational goals to guide sociologists toward the highest ideals of sociology. Although the Preamble and General Principles are not enforceable rules, they should be considered by sociologists in arriving at an ethical course of action and may be considered by ethics bodies in interpreting the Ethical Standards. The Ethical Standards set forth enforceable rules for conduct by sociologists. Most of the Ethical Standards are written broadly in order to apply to sociologists in varied roles, and the application of an Ethical Standard may vary depending on the context. The Ethical Standards are not exhaustive. Any conduct that is not specifically addressed by this Code of Ethics is not necessarily ethical or unethical. Membership in the ASA commits members to adhere to the ASA Code of Ethics and to the Policies and Procedures of the ASA Committee on Professional Ethics. Members are advised of this obligation upon joining the Association and that violations of the Code may lead to the imposition of sanctions, including termination of membership. ASA members subject to the Code of Ethics may be reviewed under these Ethical Standards only if the activity is part 221 ol or alleets llirii vvoik iclaled Iimh lions, 01 il (he ailivilv is mh inlogi. nl In nature. Personal activities having no connection to or el'leel on soi iolngisls' ,m , formance of their professional roles are not subject to the (:IHle of 1.1 In. •, PREAMBLE This Code of Ethics articulates a common set of values upon which so* iolo)>,i ,\ -, build their professional and scientific work. The Code is intended lo piovid> both the general principles and the rules to cover professional situations m countered by sociologists. It has as its primary goal the welfare and protct (ion n| the individuals and groups with whom sociologists work. It is the individual responsibility of each sociologist to aspire to the highest possible stanilaul', o| conduct in research, teaching, practice, and service. The development of a dynamic set of ethical standards for a sociologist''! work-related conduct requires a personal commitment to a lifelong effort to m i ethically; to encourage ethical behavior by students, supervisors, supervisees, employers, employees, and colleagues; and to consult with others as needed um cerning ethical problems. Each sociologist supplements, but does not violate, i lit■ values and rules specified in the Code of Ethics based on guidance drawn Irom personal values, culture, and experience. GENERAL PRINCIPLES The following General Principles are aspirational and serve as a guide for sociol ogists in determining ethical courses of action in various contexts. They exem plify the highest ideals of professional conduct. Principle A: Professional Competence Sociologists strive to maintain the highest levels of competence in their work; they recognize the limitations of their expertise; and they undertake only those tasks for which they are qualified by education, training, or experience. They recognize the need for ongoing education in order to remain professionally competent; and they utilize the appropriate scientific, professional, technical, and administrative resources needed to ensure competence in their professional activities. They consult with other professionals when necessary for the benefit of their students, research participants, and clients. Principle B: Integrity Sociologists are honest, fair, and respectful of others in their professional activities—in research, teaching, practice, and service. Sociologists do not knowingly act in ways that jeopardize either their own or others' professional welfare. Sociologists conduct their affairs in ways that inspire trust and confidence; they do not knowingly make statements that are false, misleading, or deceptive. ff I'liiK ipl<- ( : l'i olessiomil and Si lenhl'n responsibility So< iologisls adhere to (he hifhesl s> tontilic and pmlessional standards and .1« i i-pl responsibility lor theii woik. Sonologists understand that they form a ( oimmmily and show respect lor other sociologists even when they disagree on theoretical, methodological, or personal approaches to professional activities. Sociologists value the public trust in sociology and are concerned about their ethical behavior and that of other sociologists that might compromise that trust. While endeavoring always to be collegial, sociologists must never let the desire lo be collegial outweigh their shared responsibility for ethical behavior. When appropriate, they consult with colleagues in order to prevent or avoid unethical conduct. Principle D: Respect for People's Rights, Dignity, and Diversity Sociologists respect the rights, dignity, and worth of all people. They strive to eliminate bias in their professional activities, and they do not tolerate any forms of discrimination based on age; gender; race; ethnicity; national origin; religion; sexual orientation; disability; health conditions; or marital, domestic, or parental status. They are sensitive to cultural, individual, and role differences in serving, teaching, and studying groups of people with distinctive characteristics. In all of their work-related activities, sociologists acknowledge the rights of others to hold values, attitudes, and opinions that differ from their own. Principle E: Social Responsibility Sociologists are aware of their professional and scientific responsibility to the communities and societies in which they live and work. They apply and make public their knowledge in order to contribute to the public good. When undertaking research, they strive to advance the science of sociology and to serve the public good. ETHICAL STANDARDS 1. Professional and Scientific Standards Sociologists adhere to the highest possible technical standards that are reasonable and responsible in their research, teaching, practice, and service activities. They rely on scientifically and professionally derived knowledge; act with honesty and integrity; and avoid untrue, deceptive, or undocumented statements in undertaking work-related functions or activities. 2. Competence (a) Sociologists conduct research, teach, practice, and provide service only within the boundaries of their competence, based on their education, training, supervised experience, or appropriate professional experience. (Ii) Sociologists COlldlU ( l'CNCilH'll, teat Ii, pUU lief, ailll piovidf SCI vll I III III \v areas or involving new techniques only alter tliey have taken ii-asniiahlc \li p-, In ensure the competence ol their work in these areas. (c) Sociologists who engage in research, teaching, practice, or serviie m.ilniitiii awareness of current scientific and professional information in then lielif, nl activity, and undertake continuing efforts to maintain competence in tin- J-ill they use. (d) Sociologists refrain from undertaking an activity when their personal > Ii cumstances may interfere with their professional work or lead to harm Im a ,iu dent, supervisee, human subject, client, colleague, or other person to whom I In \ have a scientific, teaching, consulting, or other professional obligation. 3. Representation and Misuse of Expertise (a) In research, teaching, practice, service, or other situations where sociologist-, render professional judgments or present their expertise, they accurately ami fairly represent their areas and degrees of expertise. (b) Sociologists do not accept grants, contracts, consultation, or work assign ments from individual or organizational clients or sponsors that appear likely In require violation of the standards in this Code of Ethics. Sociologists dissociate themselves from such activities when they discover a violation and are unable to achieve its correction. (c) Because sociologists' scientific and professional judgments and actions may affect the lives of others, they are alert to and guard against personal, financial, social, organizational, or political factors that might lead to misuse of theii knowledge, expertise, or influence. (d) If sociologists learn of misuse or misrepresentation of their work, they take reasonable steps to correct or minimize the misuse or misrepresentation. 4. Delegation and Supervision (a) Sociologists provide proper training and supervision to their students, supervisees, or employees and take reasonable steps to see that such persons perform services responsibly, competently, and ethically. (b) Sociologists delegate to their students, supervisees, or employees only those responsibilities that such persons, based on their education, training, or experience, can reasonably be expected to perform either independently or with the level of supervision provided. 5. Nondiscrimination Sociologists do not engage in discrimination in their work based on age; gender; race; ethnicity; national origin; religion; sexual orientation; disability; health conditions; marital, domestic, or parental status; or any other applicable basis proscribed by law. <>. INImi exploitation Whether loi personal, et oiiomii, oi piolessional advantage, sociologists do not exploit persons ovei whom they have direct or indirect supervisory, evaluative, or other authority sikIi as students, supervisees, employees, or research participants. (b) Sociologists do not directly supervise or exercise evaluative authority over .my person with whom they have a sexual relationship, including students, supervisees, employees, or research participants. 7. Harassment Sociologists do not engage in harassment of any person, including students, supervisees, employees, or research participants. Harassment consists of a single intense and severe act or of multiple persistent or pervasive acts which are demeaning, abusive, offensive, or create a hostile professional or workplace environment. Sexual harassment may include sexual solicitation, physical advance, or verbal or non-verbal conduct that is sexual in nature. Racial harassment may include unnecessary, exaggerated, or unwarranted attention or attack, whether verbal or non-verbal, because of a person's race or ethnicity. 11. Confidentiality Sociologists have an obligation to ensure that confidential information is protected. They do so to ensure the integrity of research and the open communication with research participants and to protect sensitive information obtained in research, teaching, practice, and service. When gathering confidential information, sociologists should take into account the long-term uses of the information, including its potential placement in public archives or the examination of the information by other researchers or practitioners. 11.01 Maintaining Confidentiality (a) Sociologists take reasonable precautions to protect the confidentiality rights of research participants, students, employees, clients, or others, (bj Confidential information provided by research participants, students, employees, clients, or others is treated as such by sociologists even if there is no legal protection or privilege to do so. Sociologists have an obligation to protect confidential information, and not allow information gained in confidence from being used in ways that would unfairly compromise research participants, students, employees, clients, or others. (c) Information provided under an understanding of confidentiality is treated as such even after the death of those providing that information. (d) Sociologists maintain the integrity of confidential deliberations, activities, or roles, including, where applicable, that of professional committees, review panels, or advisory groups (e.g., the ASA Committee on Professional Ethics). (e) SlK lologislS, to lilt' extent possible, pioteil till' l I Hllldclll l.llll y III '.hhll III records, performance data, and personal inletmalion, wliediei vcihal 01 wiltu n given in die context ol academic consultation, supervision, 01 advising (f) The obligation to maintain confidentiality extends to membeis ol icm .11. Ii or training teams and collaborating organizations who have access to the intm mation. To ensure that access to confidential information is restricted, it c. tin responsibility of researchers, administrators, and principal investigators to in struct staff to take the steps necessary to protect confidentiality. (g) When using private information about individuals collected by oihei pi i sons or institutions, sociologists protect the confidentiality of individually i> l< n tifiable information. Information is private when an individual can rcasonuhh expect that the information will not be made public with personal idcnlilii r, (e.g., medical or employment records). 11.02 Limits of Confidentiality (a) Sociologists inform themselves fully about all laws and rules which urns limit or alter guarantees of confidentiality. They determine their ability to gum antee absolute confidentiality and, as appropriate, inform research participant'., students, employees, clients, or others of any limitations to this guarantee at tin outset consistent with ethical standards set forth in 11.02(b). (b) Sociologists may confront unanticipated circumstances where they become aware of information that is clearly health- or life-threatening to research pai ticipants, students, employees, clients, or others. In these cases, sociologists bal ance the importance of guarantees of confidentiality with other principles In this Code of Ethics, standards of conduct, and applicable law. (c) Confidentiality is not required with respect to observations in public places, activities conducted in public, or other settings where no rules of privacy are provided by law or custom. Similarly, confidentiality is not required in the case of information available from public records. 11.03 Discussing Confidentiality and Its Limits (a) When sociologists establish a scientific or professional relationship with persons, they discuss (1) the relevant limitations on confidentiality, and (2) the foreseeable uses of the information generated through their professional work. (b) Unless it is not feasible or is counter-productive, the discussion of confidentiality occurs at the outset of the relationship and thereafter as new circumstances may warrant. 11.04 Anticipation of Possible Uses of Information (a) When research requires maintaining personal identifiers in data bases or systems of records, sociologists delete such identifiers before the information is made publicly available. ?! |b) Win n i on hi lent ial inloi mat ii m < om i i mug i eseai i Ii pai I ii ipants, i lien Is, or oilier rei ipienls ol seiviie i*. ei 11 e 11 11 ml 11 11, it abases or systems ol records available to persons without (he pnoi i onsenl ol the relevant parties, sociologists piotei I anonymity by not including personal identifiers or by employing other techniques that mask or control disclosure of individual identities, (i ) When deletion of personal identifiers is not feasible, sociologists take reasonable steps to determine that appropriate consent of personally-identifiable individuals has been obtained before they transfer such data to others or review such data collected by others. I 1.05 Electronic Transmission of Confidential Information Sociologists use extreme care in delivering or transferring any confidential data, information, or communication over public computer networks. Sociologists are attentive to the problems of maintaining confidentiality and control over sensitive material and data when use of technological innovations, such as public computer networks, may open their professional and scientific communication to unauthorized persons. 11.06 Anonymity of Sources (a) Sociologists do not disclose in their writings, lectures, or other public media confidential, personally identifiable information concerning their research participants, students, individual or organizational clients, or other recipients of their service which is obtained during the course of their work, unless consent from individuals or their legal representatives has been obtained. (b) When confidential information is used in scientific and professional presentations, sociologists disguise the identity of research participants, students, individual or organizational clients, or other recipients of their service. 11.07 Minimizing Intrusions on Privacy (a) To minimize intrusions on privacy, sociologists include in written and oral reports, consultations, and public communications only information germane to the purpose for which the communication is made. (b) Sociologists discuss confidential information or evaluative data concerning research participants, students, supervisees, employees, and individual or organizational clients only for appropriate scientific or professional purposes and only with persons clearly concerned with such matters. 11.08 Preservation of Confidential Information (a) Sociologists take reasonable steps to ensure that records, data, or information are preserved in a confidential manner consistent with the requirements of this Code of Ethics, recognizing that ownership of records, data, or information may also be governed by law or institutional principles. (b) Sociologists plan so llnil i onlidentiality ol feioids, data, oi inloi 111 .H j < > 11 l* protected in the event ol the soc iologist's death, incapac ity, ot vvillnliav\-,il I.....i the position or practice. (c) When sociologists transfer confidential records, data, or information to i >ili< i persons or organizations, they obtain assurances that the recipients ol tin- u > ords, data, or information will employ measures to protect confidentiality at least equal to those originally pledged. 12. Informed Consent Informed consent is a basic ethical tenet of scientific research on human popu lations. Sociologists do not involve a human being as a subject in research with out the informed consent of the subject or the subject's legally aulhnii/cil representative, except as otherwise specified in this Code. Sociologists rccognl/i the possibility of undue influence or subtle pressures on subjects that may derive from researchers' expertise or authority, and they take this into account in designing informed consent procedures. 12.01 Scope of Informed Consent (a) Sociologists conducting research obtain consent from research participants or their legally authorized representatives (1) when data are collected from research participants through any form of communication, interaction, or inlet vention; or (2) when behavior of research participants occurs in a private context where an individual can reasonably expect that no observation or reporting is taking place. (b) Despite the paramount importance of consent, sociologists may seek waivers of this standard when (1) the research involves no more than minimal risk for research participants, and (2) the research could not practicably be carried out were informed consent to be required. Sociologists recognize that waivers of consent require approval from institutional review boards or, in the absence of such boards, from another authoritative body with expertise on the ethics of research. Under such circumstances, the confidentiality of any personally identifiable information must be maintained unless otherwise set forth in 11.02(b). (c) Sociologists may conduct research in public places or use publicly available information about individuals (e.g., naturalistic observations in public places, analysis of public records, or archival research) without obtaining consent. If, under such circumstances, sociologists have any doubt whatsoever about the need for informed consent, they consult with institutional review boards or, in the absence of such boards, with another authoritative body with expertise on the ethics of research before proceeding with such research. (d) In undertaking research with vulnerable populations (e.g., youth, recent immigrant populations, the mentally ill), sociologists take special care to ensure fl that the voliint.ny nalllie ol tin o m ,u< li i • iindi a stood and (hat consent is not iiii'iu'il, In all othei lespeils, siu nilogisls adheic to the principles set forth in I.'.01(a) (c). (c) Sociologists are familiar with and conform to applicable state and federal legulalions and, where applicable, institutional review board requirements for obtaining informed consent for research. 12.02 Informed Consent Process (a) When informed consent is required, sociologists enter into an agreement with research participants or their legal representatives that clarifies the nature of the research and the responsibilities of the investigator prior to conducting the research. (b) When informed consent is required, sociologists use language that is understandable to and respectful of research participants or their legal representatives. (c) When informed consent is required, sociologists provide research participants or their legal representatives with the opportunity to ask questions about any aspect of the research, at any time during or after their participation in the research. (d) When informed consent is required, sociologists inform research participants or their legal representatives of the nature of the research; they indicate to participants that their participation or continued participation is voluntary; they inform participants of significant factors that may be expected to influence their willingness to participate (e.g., possible risks and benefits of their participation); and they explain other aspects of the research and respond to questions from prospective participants. Also, if relevant, sociologists explain that refusal to participate or withdrawal from participation in the research involves no penalty, and they explain any foreseeable consequences of declining or withdrawing. Sociologists explicitly discuss confidentiality and, if applicable, the extent to which confidentiality may be limited as set forth in 11.02(b). (e) When informed consent is required, sociologists keep records regarding said consent. They recognize that consent is a process that involves oral and/or written consent. (f) Sociologists honor all commitments they have made to research participants as part of the informed consent process except where unanticipated circumstances demand otherwise as set forth in 11.02(b). 12.03 Informed Consent of Students and Subordinates When undertaking research at their own institutions or organizations with research participants who are students or subordinates, sociologists take special care to protect the prospective subjects from adverse consequences of declining or withdrawing from participation. 12.04 Informed Consent with Children (a) In undertaking research with children, sociologists obtain the i ohm nl n| children to participate, to the extent that they are capable ol providing inli consent, except under circumstances where consent may not be rctpihcd ,v, « t forth in 12.01(b). (b) In undertaking research with children, sociologists obtain the consent ol .1 parent or a legally authorized guardian. Sociologists may seek waiveis ol pit rental or guardian consent when (1) the research involves no more than mini mal risk for the research participants, and (2) the research could not pr;u tu nl-■'/ [1961 ]. Boys in White: Student Culture in Medical School. New llrunswii I, NJ: Transaction. Belgrave, Linda Liska, and Kenneth J. Smith. "Negotiated Validity in Collahoui tive Ethnography." Qualitative Inquiry 1:69-87. Berg, Bruce L. 2001. Qualitative Research Methods for the Social Sciences. Boston Allyn & Bacon. Berns, Nancy. 1999. '"My Problem and How I Solved It': Domestic Violence in Women's Magazines." The Sociological Quarterly 40:85-108. Biernacki, Patrick, and Dan Waldorf. 1981. "Snowball Sampling: Problems and Techniques of Chain Referral Sampling." Sociological Methods and Reseanli 10:141-163. Binik, Yitzchak M.; Kenneth Mah; and Sara Kiesler. 1999. "Ethical Issues in Conducting Sex Research on the Internet." Journal of Sex Research 36:82-91. Blee, Kathleen. 1996. "Becoming a Racist: Women in Contemporary Ku Klux Klan and Neo-Nazi Groups." Gender &č Society 10: 680-703. Blumer, Herbert. 1969. Symbolic Interactionism: Perspective and Method. Berkeley: University of California Press. Britton, Dana M. 1999. "Cat Fights and Gang Fights: Preference for Work in a Male-Dominated Organization." Sociological Quarterly 40:455-474. Brown, L. David. 1993. "Social Change Through Collective Reflection with Asian Nongovernmental Development Organizations." Human Relations 46:249-274. Burawoy, Michael, et al. 1991. Ethnography Unbound. Berkeley: University of California Press. Cancian, Francesca. 1996. "Participatory Research and Alternative Strategies for Activist Sociology." Pp. 187-205 in Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana: University of Illinois Press. Cannon, Lynne Weber; Elizabeth Higginbotham; and Marianne L.A. Leung. 1991. "Race and Class Bias in Qualitative Research on Women." Pp. 107-118 in Beyond Methodology: Feminist Scholarship as Lived Research, edited by Mary Margaret Fonow and Judith A. Cook. Bloomington: Indiana University Press. Caplan, Arthur L. 1992. "When Evil Intrudes." Hastings Center Report 22:29-32. 1 < hapkis, Wendy. 199/. Live Sc\ Alt'. Women I'eijotwing F.mlii Labor. New York: Roulledge. t 'Itanna/., Kathy. 2000. "(iioimdcd Theory, t )bjectivist and Construct!vist Methods." Pp. 509-536 in / landbook oj Qualitative Research, 2nd ed., edited by Norman K. Den/in and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. t bataway, Cynthia. 1997. "An Examination of the Constraints on Mutual Inquiry in a Participatory Action Research Project." Journal of Social Issues 53:747-765. (dark, Roger; Rachel Lennon; and Leanna Morris. 1993. "Of Caldecotts and Kings: Gendered Images in Recent American Children's Books by Black and Non-Black Illustrators." Gender 86 Society 7:227-245. Coffey, Amanda, and Paul Atkinson. 1996. Making Sense of Qualitative Data: Complementary Research Strategies. Thousand Oaks, CA: Sage. Cole, Caroline. 1991. "Oh Wise Women of the Stalls." Discourse and Society 2:401-411. Collins, Patricia Hill. 1990. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Routledge. Corsaro, William A., and Luisa Molinari. 2000. "Priming Events and Italian Children's Transition from Preschool to Elementary School: Representations and Actions." Social Psychology Quarterly 63:16-33. Coy, Patrick G. 1999. "Participant Observation, Shared Risks, and the Search for the Research Self Amidst Political Violence." Paper presented at the American Sociological Association annual meeting, Chicago. Davidman, Lynn. 1999. "The Personal, the Sociological, and the Intersection of the Two." Pp. 70-87 in Qualitative Sociology as Everyday Life, edited by Barry Glassner and Rosanna Hertz. Thousand Oaks, CA: Sage. Deegan, Mary Jo. 1991. Women in Sociology: A Bio-Bibliographical Sourcebook. New York: Greenwood Press. Denzin, Norman. 1989. The Research Act, 3rd ed. Englewood Cliffs, NJ: Prentice Hall. Denzin, Norman. 1999. "Performing Montana." Pp. 147-158 in Qualitative Sociology as Everyday Life, edited by Barry Glassner and Rosanna Hertz. Thousand Oaks, CA: Sage. Denzin, Norman. 2000. "The Practices and Politics of Interpretation." Pp. 897-922 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Denzin, Norman K., and Yvonna S. Lincoln. 1998. "Introduction: Entering the Field of Qualitative Research. Pp. 1-34 in The Landscape of Qualitative Research: Theories and Issues, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Denzin, Norman K., and Yvonna S. Lincoln (eds.). 2000. Handbook of Qualitative Research, 2nd ed. Thousand Oaks, CA: Sage. Deutsch, Francine M. 1999. Halving It All: How Equally Shared Parenting Works. Cambridge, MA: Harvard University Press, Do Vault., Marjorie I.. 1999. Liberating Method: Feminism ) Women ,md Girls in Poor America. Boston: Beacon Press. Dohan, Daniel, and Martin Sanchez-Jankowski. 1998. "Using Computers to An alyze Ethnographic Field Data: Theoretical and Practical Considriaiion>< Annual Review of Sociology 24:477-491. Donnelly, David F., and Kristina Ross. "The Internet: Historical Media Hcsrnu h on the Virtual Archives." Historical Journal of Film, Radio Television 17:129-137. Ellingson, Laura. 1998. '"Then You Know How I Feel': Empathy, Identification, and Reflexivity in Fieldwork." Qualitative Inquiry 4:492-513. Ellis, Carolyn. 1998. "'I Hate My Voice': Coming to Terms with Minor Bodily Stigmas." Sociological Quarterly 39:517-537. Ellis, Carolyn. 1999. "Heartful Autoethnography." Qualitative Health Resetmh 9:669-684. Ellis, Carolyn, and Arthur P. Bochner. 2000. "Autoethnography, Personal Nun a tive, Reflexivity: Researcher as Subject." Pp. 733-768 in Handbook of Quul itative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Elsinger, Heidi. 1998. "Images of Women in Sports Illustrated and Sports Illustrated for Kids." Unpublished senior thesis, University of Massachusetts-Lowell. Emerson, Robert M.; Rachel I. Fretz; and Linda L. Shaw. 1995. Writing Ethnographic Fieldnotes. Chicago: University of Chicago Press. Esterberg, Kristin. 1997. Lesbian and Bisexual Identities: Constructing Communities, Constructing Selves. Philadelphia: Temple University Press. Ferrell, Jeff. 1995. "Urban Graffiti: Crime, Control, and Resistance." Youth &■ Society 27:73-93. Fine, Michelle; Lois Weis; Susan Weseen; and Loonmun Wong. 2000. "For Whom? Qualitative Research, Representations, and Social Responsibilities." Pp. 107-131 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Finlay, William, and James E. Coverdill. 1999. "The Search Game: Organizational Conflicts and the Use of Headhunters." Sociological Quarterly 40: 11-30. Fleitas, Joan. 1998. "Spinning Tales from the World Wide Web: Qualitative Research in an Electronic Environment." Qualitative Health Research 8: 283-293. Fontána, Andrea, and James H. Frey. 1998. "Interviewing: The Art of Science." Pp. 47-78 in Collecting and Interpreting Qualitative Materials, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Ii a n /osi, Hol hi I o I 990a. "Si i nl« -gl« "• h>i I he Pi even! ion, I )ele« lion, and ( oi lection ol Measuirmi-nl |''.imi m I >ai:i < olln led IromTeNliial Sources." Socio loginil Methods and Kesnm h I HAM \ 12. Iian/.osi, Roberto. 1990k "(lomputor-Assisled Coding of Textual Data: An Application to Semantic Grammars." Sociological Methods and Research 19:225-257. I Yan/.osi, Roberto. 1998. "Narrative Analysis—Or Why (and How) Sociologists Should Be Interested in Narrative." Annual Review of Sociology 24:517-554. l'reedman, Carol. 1997. "Setting Stay-at-Home Standards: An Ethnographic Study of the Southland Mothers Association." Unpublished M.A. thesis, University of Missouri-Kansas City. Freire, Paulo. 1970. Pedagogy of the Oppressed. New York: Seabury Press. Friedman, Debra, and Michael Hechter. 1988. "The Contribution of Rational Choice Theory to Macrosociological Research." Sociological Theory 6: 201-218. Gamson, Joshua. 1998. Freaks Talk Back: Tabloid Talk Shows and Sexual Nonconformity. Chicago: University of Chicago Press. Gans, Herbert J. 1962. The Urban Villagers: Group and Class in the Life of Italian-Americans. New York: Free Press. Gans, Herbert J. 1999. "Participant Observation in the Era of 'Ethnography.'" Journal of Contemporary Ethnography 28:540-548. Ganz, Marshall. 2000. "Resources and Resourcefulness: Strategic Capacity in the Unionization of California Agriculture, 1959-1966." American Journal of Sociology 105:1003-1062. Garber, Marjorie. 1995. Vice Versa: Bisexuality and the Eroticism of Everyday Life. New York: Simon & Schuster. Gaventa, John. 1991. "Carrying On . . ." Social Policy 21 (3):68-70. Gee, James. 1991. "A Linguistic Approach to Narrative." Journal of Narrative and Life History 1:15-39. Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books. Gergen, Kenneth J. 1991. The Saturated Self: Dilemmas of Identity in Contemporary Life. Basic Books. Gergen, Mary M., and Kenneth J. Gergen. 2000. "Qualitative Inquiry: Tensions and Transformations." Pp. 1025-1046 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Gibson-Graham, J.K. 1994. '"Stuffed If I Know1.': Reflections on Post-modern Feminist Social Research." Gender Place Sč Culture: A Journal of Feminist Geography 1:205-224. Glassie, Henry. 1991. "Studying Material Culture Today." Pp. 253-266 in Living in a Material World: Canadian and American Approaches to Material ('ul ture, edited by Gerald Pocius. St. John's, Newfoundland: Institute of Social and Economic Research. 1999. ()itiilil,nii'c Soiiolofti ,ia i'iviY .ondon: Macmillan. 7'/((.• Politics and I lisiorv ol lam Glassner, Harry, ami Rosanna 1 lerlz (ed.« day Life. Thousand Oaks, (.-A: Sage Goffman, Erving. 1976. Gender Advertisements. Gordon, Linda. 1988. Heroes of Their Own Lives ily Violence. New York: Penguin Books. Gottlieb, Alma, and Philip Graham. 1993. Parallel Worlds: An Anthropologist iind a Writer Encounter Africa. New York: Crown, 1993. Greenwood, Davydd J.; William Foote Whyte; and Ira Harkavy. 1993."PartU Ipu tory Action Research as a Process and as a Goal." Human Relation* 46:179-193. Greer, Colin. 1991. "A Culture of Politics." Social Policy 21(3):53-57. Groger, Lisa; Pamela Mayberry; and Jane Straker. 1999. "What We Didn't Leant Because of Who Would Not Talk to Us." Qualitative Health Rosea hh 9:829-836. Gubrium, Jaber F., and James A. Holstein. 1997. The New Language of Quctlltit tive Method. New York: Oxford University Press. Gubrium, Jaber R, and James A. Holstein. 1998. "Narrative Practice and the ("o- herence of Personal Stories." Sociological Quarterly 39:163-187. Harding, Sandra. 1986. The Science Question in Feminism. Ithaca, NY: Cornell University Press. Harding, Sandra. 1987. "Introduction: Is There a Feminist Method?" Pp. 1-14 li\ Feminism Sd Methodology, edited by Sandra Harding. Bloomington: Indiaiui University Press. Hartsock, Nancy. 1987. "The Feminist Standpoint: Developing the Ground for a Specifically Feminist Historical Materialism." Pp. 157-180 in Feminism &' Methodology, edited by Sandra Harding. Bloomington: Indiana University Press. Hays, Sharon. 1996. The Cultural Contradictions of Motherhood. New Haven, CT: Yale University Press. Hodder, Ian. 1989. "Post-modernism, Post-structuralism and Post-processual Archaeology." Pp. 64-78 in The Meanings of Things: Material Culture and Symbolic Expression, edited by Ian Hodder. London: Unwin Hyman. Hodder, Ian. 1998. "The Interpretation of Documents and Material Culture." Pp. 110-129 in Collecting and Interpreting Qualitative Materials, edited by Norman K. Denzin and Yvonna S. Lincoln.Thousand Oaks, CA: Sage. Holstein, James A., and Jaber F. Gubrium. 1995. The Active Interview. Thousand Oaks, CA: Sage. Hondagneu-Sotelo, Pierrette. 1996. "Immigrant Women and Paid Domestic Work: Research, Theory, and Activism." Pp. 105-122 in Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana: University of Illinois Press. Huberman, A. Michael, and Matthew B. Miles. 1998. "Data Management and Analysis Methods." Pp. 179-210 in Collecting and Interpreting Qualitative Materials, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Humphreys, Laud. 19/(1. 'Icawoin lUule: Impersonal Se.\ in Public Places. (Jti-iago: Aldine. Ignacio, Emily Noelle. 2000. "'Ain't I a Filipino (Woman)?' An Analysis of Authorship/Authority Through the Construction of Tilipina' on the Net." Sociological Quarterly 41:551-572. lanesick, Valerie. 1998. "Stretching" Exercises for Qualitative Researchers. Thousand Oaks, CA: Sage. lanis, Irving. 1982. Groupthink: Psychological Studies of Policy Decisions and Fias-coes. Boston: Houghton Mifflin. Jayaratne, Toby Epstein, and Abigail J. Stewart. 1991. "Quantitative and Qualitative Methods in the Social Sciences: Current Feminist Issues and Practical Strategies." Pp. 85-106 in Beyond Methodology: Feminist Scholarship as Lived Research, edited by Mary Margaret Fonow and Judith A. Cook. Bloomington: Indiana University Press. Karp, David A. 1999, "Social Science, Progress, and the Ethnographer's Craft." Journal of Contemporary Ethnography 28:597-609. Katz, Jack. 1996. "Families and Funny Mirrors: A Study of the Social Construction and Personal Embodiment of Humor." American Journal of Sociology 101:1194-1237. Kendall, Lori. 2000. '"OH NO! I'M A NERD!' Hegemonic Masculinity on an Online Forum." Gender & Society 14:256-274. Keough, Noel; Emman Carmona; and Linda Grandinetti. 1995. "Tales from the Sari-Sari: In Search of Bigfoot." Convergence 28(4):5-16. Kincheloe, Joe L., and Peter L. McLaren. 1998. "Rethinking Critical Theory and Qualitative Research." Pp. 260-299 in The Landscape of Qualitative Research: Theories and Issues, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Kleinman, Sherryl. 1991 ."Field-Workers' Feelings: What We Feel, Who We Are, How We Analyze." Pp. 184-195 in Experiencing Fieldwork: An Inside View of Qualitative Research. Newbury Park, CA: Sage. Koester, Stephen. 1994. "Copping, Running, and Paraphernalia Laws: Contextual Variables and Needle Risk Behavior Among Injection Drug Users in Denver." Human Organization 53:287-295. Kondo, Dorinne K. 1990. Crafting Selves: Power, Gender, and Discourses of Identity in a Japanese Workplace. Chicago: University of Chicago Press. Kotamraju, Nalini P. 1999. "The Birth of Web Site Design Skills: Making the Present History." American Behavioral Scientist 43:464-475. Krieger, Susan. 1983. The Mirror Dance: Identity in a Women's Community. Philadelphia: Temple University Press. Krieger, Susan. 1991. Social Science and the Self: Personal Essays on an Art Form. New Brunswick, NJ: Rutgers University Press. Johnson, Timothy P.; Michael Fendrich; Chitra Shaligram; Anthony Garcy; and Samuel Gillespie. 2000. "An Evaluation of the Effects of Interviewer Characteristics in an RDD Telephone Survey of Drug Use." Journal of Drug Issues 30:77-102. Hel ween Stu inlogy .Ulil l.lngnlMii ., I.abov, William. 1978. "( iossing (lie (in American Sociologist 13:93 103. Lamott, Anne. 1995. Bird by Bird: Some Instructions on Writing mid /.i/c N> w York: Anchor. Lareau, Annette, and Jeffrey Schultz (eds.j. 1996. .tourneys Through l:thnoghtin ml< Analysis." Journal of Marriage and the Family 62:375-388. Laslett, Barbara, and Barrie Thorne. 1997. Feminist Sociology: Life Histories n( a Movement. New Brunswick, NJ: Rutgers University Press. Leblanc, Lauraine. 2000. Pretty in Punk: Girls' Gender Resistance in a liirys' Sob culture. New Brunswick, NJ: Rutgers University Press. LeCompte, Margaret D., and Jean J. Schensul. 1999a. Designing and Comlih ini>\ Ethnographic Research. Walnut Creek, CA: AltaMira Press. LeCompte, Margaret D., and Jean J. Schensul. 1999b. Analyzing and Interpreting Ethnographic Data. Walnut Creek, CA: AltaMira Press. Lee, Raymond M. 2000. Unobtrusive Methods in Social Research. Buckingham, England: Open University Press. Lempert, Lora Bex. 1994. "A Narrative Analysis of Abuse." Journal of Contempt rary Ethnography 22:411 -441. Lerner, Jennifer E,, and Linda Kalof. 1999. "The Animal Text: Message and Meaning in Television Advertisements." The Sociological Quarterly 40. 565-586. Liebow, Elliot. 1967. Tally's Comer: A Study of Negro Streetcorner Men. Boston: Little, Brown. Liebow, Elliot. 1993. Tell Them Who I Am: The Lives of Homeless Women. New York: Penguin Books. Lincoln, Yvonna S., and Norman K. Denzin. 2000. "The Seventh Moment: Out of the Past." Pp. 1047-1065 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Lincoln, Yvonna S., and Egon Guba. 1985. Naturalistic Inquiry. Beverly Hills, CA: Sage. Lincoln, Yvonna S., and Egon Guba. 2000. "Paradigmatic Controversies, Contradictions, and Emerging Confluences." Pp. 163-188 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage Lofland, John, and Lyn H. Lofland. 1984. Analyzing Social Settings, 2nd ed. Belmont, CA: Wadsworth. Lofland, John, and Lyn Lofland. 1995. Analyzing Social Settings, 3rd ed. Belmont, CA: Wadsworth. Lofland, Lyn. 1985. A World of Strangers: Order and Action in Urban Public Space. Prospect Heights, IL: Waveland Press. I.ykes, M Itiinloii. I' I' I /. "Ai 11 vr.t I'.ntii i|>aiiitv Keseauli Among (he Maya of t iiialeinala: ( 'onsli u> I mg Mi-an mg;, I nun Silualed I .angtiagc." Journal nj So rial Issues 51:725 1 \ I. Mat donalcl, Keith, and (lolin Tipton. 1993. "Using Documents." Pp. 187-200 in Researching Social Lije, edited by Nigel Gilbert. London: Sage. MacDougall, Colin, and Frances Baum. 1997. "The Devil's Advocate: A Strategy to Avoid Groupthink and Stimulate Discussion in Focus Groups." Qualitative Health Research 7:532-540. Madriz, Esther I. 1998. "Using Focus Groups with Lower Socioeconomic Status Latina Women." Qualitative Inquiry 4:114-129. Madriz, Esther. 2000. "Focus Groups in Feminist Research." Pp. 835-850 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Manning, Peter K., and Betsy Cullum-Swan. 1998. "Narrative, Content, and Semiotic Analysis." Pp. 246-273 in Collecting and Interpreting Qualitative Materials, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Marcenko, Maureen O, and Linda Samost. 1999. "Living with HIV/AIDS: The Voices of HIV-Positive Mothers." Social Work 44:36-45. Marcus, George. 1998. "What Comes (Just] After 'Post? The Case of Ethnography." Pp. 383-406 in The Landscape of Qualitative Research: Theories and Issues, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage Masuda, Kazuyo. 1998. "Cockroaches and Cops: Women Struggling in an Urban Section 8 Housing Complex." Unpublished M.A. thesis, University of Missouri-Kansas City. McAdams, Dan P. 1993. The Stories We Live By: Personal Myths and the Making of the Self. New York: Guilford Press. McNeil, Larry B. 1995. "The Soft Arts of Organizing." Social Policy 26:16 72. McNicolI, Paule. 1999. "Issues in Teaching Participatory Action Research." loin nal of Social Work Education 35:51-62. Miles, Matthew B., and A. Michael Huberman. 1994. Qualitative I )ata /Iim/vw. An Expanded Sourcebook. Thousand Oaks, CA: Sage. Milgram, Stanley. 1974. Obedience to Authority. New York: llaipei & How Miller, Neil. 1995. Out of the Past: Gay and Lesbian History \iotn IHi<'> to the Present. New York: Vintage Books. Mishler, Elliot G. 1991. "Representing Discourse: The Kluioiii olTnne.i ii|.u>.n Journal of Narrative and Life History 1:255-28(). Mitchell, Richard G., Jr. 1991. "Secrecy and Disclosine In I i. I.■ ■!I I'p 97-108 in Experiencing Fieldwork: An Inside \ 'ietv o\ ijiialnailri lU-niťU Newbury Park, CA: Sage. Mitchell, Richard G. 1993. Secrecy and ftcldwoik. Niwbui y I'm !•, < A 'i.t|» Moen, Phyllis. 1989. Working Parents: Transitu inaiioin m < mmi/ii /Inli-n itiiif / n/ lie Policies in Sweden. Madison: University ol WIm omin I'n •<■« MonU'll, Frances. I !>'>'). "I'm us (.lump Interviews: A New Feminist Met Im. I " NWS A Journal I 1:4 1 /I. Montini, Theresa. 1996. "(lender and Emotion in the Advocacy lor Hicasl ( an cer Informed Consent Legislation." Gender &C Society 10:9-23. Morgan, David L. 1988. Focus Croups as Qualitative Research. Newbmy I*.111>, CA: Sage. Morgan, David L. 1995. "Why Things (Sometimes) Go Wrong in Focus (iioiips" Qualitative Health Research 5:516-524. Morgan, David L. 1996. "Focus Groups." Annual Review of Sociology 129-153. Morse, Janice M. 1998. "The Contracted Relationship: Ensuring Protection ol Anonymity and Confidentiality." Qualitative Health Research 8:301-304. Morse, Janice M. 1998. "Designing Funded Qualitative Research." Pp. 56-85 in Strategies of Qualitative Inquiry, edited by Norman K. Denzin and Yvonnn S Lincoln. Thousand Oaks, CA: Sage. Naples, Nancy A., with Emily Clark. 1996. "Feminist Participatory Research and Empowerment: Going Public as Survivors of Childhood Sexual Abuse." Pp. 160-183 in Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana: University of Illinois Press. Neuman, W. Lawrence. 1994. Social Research Methods, 2nd ed. Boston: Allyn & Bacon. Neutens, James J., and Laura Rubinson. 1997. Research Techniques for the Health Sciences, 2nd ed. Boston: Allyn & Bacon. Nielsen, Joyce McCarl. 1990. Feminist Research Methods: Exemplary Readings in the Social Sciences. Boulder, CO: Westview Press. OakleyAnn. 1981."InterviewingWomen:A Contradiction inTerms."Pp. 30-61 in Doing Feminist Research, edited by Helen Roberts. London: Routledge 8i Kegan Paul. Park, Robert E. On Social Control and Collective Behavior. Chicago; University of Chicago Press, 1967. Patton, Michael Quinn. 1990. Qualitative Evaluation and Research Methods, 2nd ed. Newbury Park, CA: Sage. Plummer, Ken. 1995. Telling Sexual Stories. London: Routledge. Plummer, Ken. 1999. "The 'Ethnographic Society' at Century's End." Journal of Contemporary Ethnography 28:641 -649. Posavac, Emil J., and Raymond G. Carey. 1997. Program Evaluation: Methods and Case Studies, 5th ed. Upper Saddle River, NJ: Prentice Hall. Presser, Stanley, and Linda Stinson. 1998. "Data Collection Mode and Social Desirability Bias in Self-Reported Religious Attendance." American Sociological Review 63:137-145. Punch, Maurice. 1998. "Politics and Ethics in Qualitative Research." Pp. 156-184 in The Landscape of Qualitative Research, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Ragin, < liai les < '.. I 994 < onstiiii /(»>; .'.aks, ( '.A: Pine Foige Pi ess. Kalhjc, William. 1992. "I low MikIi Alcohol I >o We Drink? It's a Question . . . Solo Speak." Gk, ('aiol. 19/4. All < >ki Kin '*tialette*. \oi Survival in a Bhuk Community. New York: I larpei Si ld>\v Slack, Carol. 1996. "Willing Ethnography: Feminist Critical Practice." Pp. 96 106 in Feminist hileinmas in Fieldwork, edited by Diane L. Wolf Boulder, CO: Westview. Steinberg, Ronnie J. 1996. "Advocacy Research for Feminist Policy Objectives: Experiences with Comparable Worth." Pp. 225-255 in Feminism and Social Change: Bridging Theory and Practice, edited by Heidi Gottfried. Urbana: University of Illinois Press. Stoecker, Randy. 1999. "Are Academics Irrelevant? Roles for Scholars in Participatory Research." American Behavioral Scientist 42:840-844. Strauss, Anselm, and Juliet Corbin. 1990. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park, CA: Sage. Strauss, Anselm, and Juliet Corbin. 1998. "Grounded Theory Methodology: An Overview." Pp. 158-183 in Strategies of Qualitative Inquiry, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Stringer, Ernest T. 1996. Action Research: A Handbook for Practitioners. Thousand Oaks, CA: Sage. Tewksbury, Richard, and Patricia Gagne. 1997. "Assumed and Presumed Identities: Problems of Self-Presentation in Field Research." Sociological Spectrum 17:127-156. Theophano, Janet, and Karen Curtis. 1996. "Reflections on a Tale Told Twice." Pp. 151-176 in Journeys Through Ethnography: Realistic Accounts of Field-work, edited by Annette Lareau and Jeffrey Schultz. Boulder, CO: West-view Press. Thompson, Sharon. 1995. Going All the Way: Teenage Girk' Tales of Sex, Romance, and Pregnancy. New York: Hill & Wang. Thorne, Barrie. 1993. Gender Play: Girls and Boys in School New Brunswick, NJ: Rutgers University Press. Tuchman, Gaye. 1998. "Historical Social Science: Methodologies, Methods, and Meanings." Pp. 225-260 in Strategies of Qualitative Inquiry, edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Viditch, Arthur J., and Stanford M. Lyman. 2000. "Qualitative Methods: Their History in Sociology and Anthropology." Pp. 37-84 in Handbook of Qualitative Research, 2nd ed., edited by Norman K. Denzin and Yvonna S. Lincoln. Thousand Oaks, CA: Sage. Walters, Ian. 1997. "Vietnam Zippos." Journal of Material Culture 2:61-75. Walzer, Susan. 1998. Thinking About the Baby: Gender and Transitions into Parenthood. Philadelphia: Temple University Press. Webb, Eugene J.; Donald T. Campbell; Richard Schwartz; Lee Sechrest; ami Janet Belew Grove. 1981. Nonreactive Measures in the Social Sciences, 2nd ed. Boston: Houghton Mifflin. Werner, Oswald, and Mark G. Schoepfle. 1987. Systematic Fieldwork: Ethnographic Analysis and Data Management, Volume 2. Newbury Park, CA: Sage. Whittior, Nancy. 1995. Feminist (ieneralions: I'lie Persistence «I the R,i,IU,il Women's Movement. Philadelphia: Temple University Piess. Whyte, William Foote. 15)43. Street Corner Society. Chicago: Univeisily oil hi cago Press. Whyte, William Foote. 1991. "The Social Sciences in the University."/bur; it l OllMl.s llllll'llll, HI I J-' chain irlrinil «iMti)i|iMy **• ll""l IM I I HI ii ..It,inutim tWMMWft (Mitt etJMHl .min.ut**hlHi44 Oi m M IK coding, 157 162 Coffey, Amanda, 158, 166, 180, 197 Cole, Caroline, 36 collaborative research, 49, 144 Collins, Patricia Hill, 12, 19 coming-out stories, 189 communication, nonverbal. See nonverbal communication complication, in narrative analysis, 184, 193 computer programs for qualitative data analysis, 161, 162, 177-179 computers, in qualitative research, 176-179 Comte, Auguste, 10 confidentiality, 43, 45, 50-54,111,225-228 Constructing Social Research (R agin), 134 content analysis, 171-172 content latent, 172 manifest, 172 Convergence, 149 conversation analysis, 108, 184 conversation, turn taking in, 85 Corbin, Juliet, 180 Cornell University, 130 cover story, developing, 70-71 Coverdill, James, 84 covert research, 66, 71,80, 230 Crafting Selves (Kondo), 24 Creswell, John, 41 Criminal Justice Abstracts, 38 Cullum-Swan, Betsy, 173, 180 culture, 58-59 Curtis, Karen, 65 data analysis as process of making meaning, 79-80, 152-153 beginning, 151 computer-assisted, 176-179 grounding, 173-176 stages in, 151 IS'.' data i olli'i (iitii in m linn research, I 'Mi, 144 See also specific methods of data collection data complication, 166 data management, 153-157 data organizing, 154-155 data reduction, 166 data sharing, 232-233 Davidman, Lynn, 26 deception, in social research, 47-48, 52, 230 deductive reasoning, 6, 8, 10 Deegan, Mary Jo, 8 definition of the situation, 28 Denzin, Norman K., 24, 216 description, in field notes, 74-75 Designing and Conducting Ethnographic Research (LeCompte and Schensul), 41 DeVault, Marjorie, 24 devil's advocate, in focus groups, 111 diagrams, 162-164 Dickson, Geraldine, 142 documents, 121-122 Dohan, Daniel, 180 Dürkheim, Emile, 5 Eastman, Crystal, 139 editing, 203-204 Ellingson, Laura, 12 Emerson, Robert, 73, 82 emotions, in field research, 72. See also research topics, feelings about empowerment, in participatory action research, 141-142 ERIC, 38 erosion, measures of, 117 ethical codes. See codes of ethics ethical dilemmas in qualitative research, 44, 47, 51 in covert research, 53 in focus groups, 111 in Internet research, 128 in participant observation, 67-68 in :u liim irscait li, MS 1-1«» ethics. "12. IU, .|H ■(<> The I'.tlinny.rapll, 178 Ethnographer's Toolkit, I I ethnographers, 8 Ethnography Unbound (Hum woy ct al.), 23, 82 ethnography. 8, 51, 58 (>() See also participant observation evaluation research, 140 Experiencing Fieldwork (Shaffir and Stebbins, eds.), 68. 82 exploitation, of research participants, 78-79 face sheet, 101 Feminist Dilemmas in Field ■ work (Wolf), 56 Feminist Methods in Social Research (Reinharz), 150 Ferrell, Jeff, 36, 55 fiction, in social research, 21, 187,216-217 field notes avoiding generalizations in, 76 description in, 75 direct quotations in, 77 formats for, 77-78 full, 74 in interviews, 107 contents of, 74-77 when to take, 73 field research, 58. See abo ethnography, participant observation field researcher, 13. See also ethnographers field settings. See field sites field sites boundaries of, 63 evaluating, 62, 67-68 familiarity with, 64-65 gaining access to, 65-67, 68-69, 81 leaving, 78-79 selecting, 62-73 field studies, 58. See also ethnography, participant observation Fine, Michelle, 56 Finlay, William, 83 lli-it;i\ I.Kill, l."> /in ii.s ( iioups ,i\ (.tiiii/ldiHiv Kcsciinh (Morgan), I 14 Incus groups advantages of, 109 ethical issues in, 111 number of participants in, 110 role of, 108-110 tape recording, 111 focused coding, 161-162, 166-167 Fontana, Andrea, 113 Freaks Talk Back (Gamson), 124 Freedman, Carol, 26 Freire, Paulo, 137 Fretz, Rachel, 73, 82 Frey, James H., 113 friendship, in social research, 51,78 Gagne, Patricia, 90 Gamson, Joshua, 124, 209 Ganz, Marshall, 129 Garbage Project, 115, 116 Garber, Marjorie, 190 gatekeepers, 66-67 Geer, Blanche, 23 Geertz, Clifford, 16 Gergen, Kenneth, 21 Gibson-Graham, J. K., 139 Glassner, Barry, 27, 41 Gordon, Linda, 122 Gottlieb, Alma, 58 graffiti, 36, 55,116-117 Graham, Philip, 58 Grandinetti, Linda, 138 Groger, Lisa, 94 grounded research, 7, 8 grounded theory, 34 group dynamics, 110, 148 group interviews, 108-112. See also focus groups groupthink, 111 Guba, Egon, 24 Gubrium, Jaber, 2,113, 188, 194 Guide to Writing Sociology Papers (Sociology Writing Group), 219 Handbook of Qualitative Research (Denzin and Lincoln, eds.), 24 I Iniillni;. '.nulla, I ' hill in, litlnlml/th^, V> I I.iiI-.mi I., Nam v. 12 I lays, Mianm, 3t-\iyn (< h-aw-II), II H'.simi. Ii lai illlator, M 1, III research participants, 88, 93-94. See also informants research process as social production, 16 stages of, 5-7, 39-40 research project, getting started on, 25 research questions, 27, 29-30, 31 research reports, 207-215 research strategies developing, 33-37, 40 types of, 35-37 research subjects. See research participants research topics developing, 28-31 evaluating, 30-33, 39 feelings about, 28-29, 30 generating, 26-28, 39 refining, 25, 29-30 turning into research question, 29-30 research traditions, 33-35. See also paradigms research basic, 4, 138 benefits of, 32-33 covert. See covert research critical, 10, 13, 17-18,31, 34 developing relationships in, 90 evaluation, 140 feminist, 10, 13, 17, 18-20,31,51, 139-140 interpretive, 15-17 outcomes of, 135 positivist. See positivism postmodernist, 31,216. See also postmodernism quantitative versus qualitative, 2 reasons for conducting, 3-4 social constructionist, 31. See also social constructionism lesoilli es, 111) l|ll,llll.lU\e leseaii h, I.' results, in narrative analysis, 184,193 revision, 203-204 revision checklist, 205 Richardson, James, 52, 56 Richardson, Laurel, 21, 187, 215,219 Riessman, Catherine Kohler, 192, 195, 197 risks, in participant observation, 68 role taking, in field research, 69, 70 Sacks, Harvey, 184 sampling methods chain referral. See sampling methods, snowball purposive, 93 snowball, 93-94 Sanchez-Jankowski, Martin, 180 Scarce, Rik, 43 Schensul, Jean, 41, 153 Schiffman, Josepha, 209 Schultz, Jeffrey, 41 scientific method, 1, 5 search services, on-line, 37 Sechrest, Lee, 134 secondary sources, 123, 129 Seidman, Irving, 114 self as research instrument, 61 in relation to field site, 64-65 in social research, 12-13 self-disclosure, in field research, 71 semiotic analysis, 172-173 settlement movement, 8 Shaffir, William, 58, 71,82 Shaw, Linda, 73, 82 Shultz, Jeffrey, 82 Small, Stephen, 150 Smith, Dorothy, 11,19 Smith, Phil, 187, 197 snowball sampling, 93-94 social change, as outcome in action research, 146-147 social constructionism, 10, 13, 15-17,34,35 social desirability bias, 8(i social interaction, 1 5, 28 social reality, interpretation of, 16 Social Research Methods (Neuman), 134 social research, feminist critiques of, 20 Society in America (Mar- tineau), 8 Sociological Abstracts, 37, 38 sociological imagination, 4 Sociology Writing Group, 201,219 sociology, transformation of, 19 Sohrabi, Nader, 129 sources primary, 123, 129, 130-131 secondary, 123, 129 Stacey, Judith, 24, 51, 54, 56, 78, 93 Stack, Carol, 24, 90 stakeholders, identifying, 142-143 Stebbins, Robert, 71, 82 Steinberg, Ronnie, 4 Stoecker, Randy, 141, 144 stories. See narratives Stories We Live By, The (McAdams), 197 Straker, Jane, 94 Strauss, Anselm, 23, 180 Street Corner Society (Whyte), 24 "Stretching" Exercises for Qualitative Researchers (Janesick}, 72, 82,218 Stringer, Ernest, 135, 141, 142, 150 Style Guide of the American Sociological Association, 214 sin lace meaning, 132 survey research, 85 Svvartz, Richard, I 34 Symbolic Interaction, 23 symbolic interactionism, 15 Systematic Pieldwork (Werner and Scheopfle), 156 Tally's Corner (Liebow), 13, 24 tape recording, interviews, 106- 107, 111 Tell Them Who I Am (Liebow], 26 texts analysis of, 36, 121-128, 171-172 electronic, 125-128 Theophano, Janet, 65 theory reconstruction, 9, 23 theory, in social research, 4 thick description, 50 Thompson, Sharon, 108 transcription of tapes, 107- 108, 192-193 triangulation, 36-37, 176 Tricks of the Trade (Becker), 175, 180 Tuchman, Gaye, 128, 134 Tuskegee study, 45-46 typologies, 169-170 University of Maryland Broadcast Pioneers Library, 127 Unobtrusive Measures (Webb et al.), 134 Unobtrusive Measures in Social Research (Lee), 134 imolid uslve meiism es, Id I I 5. Sec