Qualitative Psychology
or the assumption of a unitary .scientific method of observation, experimentation, logic, and evidence, Positivistic beliefs in scientific logic, objectivity, and truth supported and legitimized reducing qualities of human experience to quantifiable variables. Positivistic methods assumed an unbiased and passive observer, the separation of fae! from value, the existence of an external world separate Írom scientific observers and their methods, and the accumulation of knowledge about this world. Hence, positivism led to a quest for valid instruments, replicabie research designs, and reliable findings. Most quantitative methodologists of the 1960s ignored human problems that did not fit positivistic research designs, if they acknowledged qualitative research at all, proponents of quantification considered it to be a preiiminary exercise for refining quantitative instruments. Simultaneously, the division between theory and research grew. At that time, theory informed quantitative research through the logico-deductive model of inquiry, which relied on deducing testable hypotheses from an existing theory. Yet this research seldom led to new theory construction.
In their initial statement of grounded theory, Glaser and Strauss (1967) challenged:
•     the arbitrary division of theory and research
«    prevailing views of qualitative research as a precursor to more 'rigorous' quantitative methods
•     beliefs that qualitative methods were impressionistic and unsystematic o    separation of data collection and analysis phases of research
•     assumptions that qualitative research could not generate theory.
Glaser and Strauss built on their qualitative predecessors' implicit analytic procedures and research strategies and made them explicit. Earlier qualitative researchers had taught generations of students through mentoring and immersion in field experience {Rock, 1979). Glaser and Strauss's written guidelines for conducting qualitative research changed that oral tradition. The epistemologica! assumptions, logic, and systematic approach of grounded theory methods reflect Glaser's rigorous quantitative training at Columbia University. Strong links to symbolic interaction, with its stress on human reflection, choice, and action, stem from Strauss's training at the University of Chicago with Herbert Blurner and Robert Park. Through their influence, Strauss adopted both the pragmatist philosophical tradition with its emphasis on studying process, action, and meaning (Blumer, 1969; Mead, 1934) and the Chicago legacy of ethnographic research (Park and Burgess, 1921).
Grounded theory      85
Grounded theory contains both positivistic and interpretive elements, lis emphasis on using systematic techniques to study an external world remains consistent with positivism. Its stress on how people construct actions, meanings, and intentions is in keeping with interpretive traditions. Some grounded theorists join me (see, for example, Clarke, 1998) in assuming that a researcher's disciplinary and theoretical proclivities, relationships and interactions with respondents all shape the collection, content, and analysis of data. Grounded theory can bridge traditional positivistic methods with interpretive methods in disciplines such as psychology that have embraced quantification. These methods allow psychologists to study aspects of human experience that remain inaccessible with traditional verification methods. The grounded theory emphasis on process enables psychologists to study how individual and interpersonal processes develop, are maintained, or change.
Generating Data
With grounded theory, you begin by exploring general questions about a research topic of interest. You collect data about what people who have relevant experience of this topic say and do about it. Grounded theorists' background assumptions and disciplinary interests alert them to look for certain issues and processes in their data. Consistent with Blumer's (1969) depiction of 'sensitizing concepts', grounded theorists often begin their studies with certain research interests and a set of general concepts. These concepts give you ideas to pursue and sensitize you to ask particular kinds of questions about your topic. For example, ! began my studies of people with chronic illnesses with an interest in how they experienced time and how their experiences of illness affected them. My guiding interests brought concepts such as self-concept, identity, and duration into the study. But that was only the start. 1 used those concepts as points of departure to form interview questions, to look at data, to listen to interviewees, and to think analytically about the data. Guiding interests should provide you with such points of departure for developing, rather than limiting, your ideas. 7'hcn you develop specific concepts by examining your ideas through successive stages of analysis and studying your data.
Thus, sensitizing concepts provide a place to start, not end. A thorough foundation in a discipline provides such concepts. Professional researchers already hold epistemologica! assumptions about the world, disciplinary perspectives, and often an intimate familiarity with the research topic and the pertinent literature. Yet every grounded theory researcher should remain as open as possible to new views during the research.
Hence, grounded theorists develop their sensitizing concepts in relation to the processes they define in their data. In contrast, the logico-deductive
Qualitative Psychoiogy
Grounded theory      87
model of traditional model of research necessitates operalionaiizing the previously established concept as accurately as possible.
In grounded theory research, you begin analysing what you gather early in your data collection. Simultaneous involvement in data collection and analysis means that your emerging analysis shapes your data collection decisions. Early analytic work leads you lo collect more data around emerging themes and questions. For example, we sense Susan Nelson's efforts to account for her pain and fatigue in the interview excerpt above. Her remarks alert the interviewer to ask how she discovered her other conditions and how other people responded to both her search and her conclusions. Then, further questions may be built into subsequent interviews with other participants.
Through simultaneous involvement in data collection and analysis, you avoid being overwhelmed by volumes of general, unfocused data that do not lead lo anything new. If you already have collected a substantial amount of data, begin with it, but subsequently collect additional data about your emerging analytic interests and themes. That way, you can follow up topics that are explicit in one interview or observation but remain implicit or absent in others. For example, a woman with multiple sclerosis mentioned having 'bad days'. She said, '! deal with time differently (during a bad day when she felt sick] and time has a different meaning lo me' {Charmaz, 1991a: 52). When we discussed meanings of time, 1 saw how she connected experiencing time with images of self. On a bad day, her day shortened because all her daily routines - such as bathing, dressing, exercising, and resting - lengthened substantially. As her daily routines stretched, her preferred self shrunk- After I saw how she defined herself in relation to mundane daily routines, I asked interview questions that directly addressed this relationship. Her comment provided a valuable source of comparison, along with ideas to corroborate in other interviews. For example, this piece of data aliowed me to frame new questions. To what extent do people view themselves as separated from or embedded in their daily routines? Which daily routines? How does sickness affect their views? When do they claim the self that they experience while ill? When do they reject it?
The core components of grounded theory studies are analytic categories developed while studying the data rather than preconceived concepts or hypotheses. These categories move your study toward abstract analyses yet simultaneously elucidate what happens in the empirical world.
From the beginning, researchers actively construct their data with study participants. The first question to ask is, 'What is happening here?' (Glaser, 1978, 1992; Glaser and Strauss, 1967). Then you have to think of ways to find out. Perhaps their enthusiasm for developing an inductive methodology that anchored emergent theory in data led Glaser and Strauss (1967; Glaser, 1978) to imply in their early works that categories inhere in the data and may even leap out. 3 disagree. Rather, categories reflect interactions between
the observer and observed. Certainly, social researchers' world-views, disciplinary assumptions, theoretical proclivities and research interests shape what f hey create (see also Dey, 1999) by influencing their observations and emerging categories. What happens if the data do not illuminate your initial interests? Grounded theorists evaluate the fit between their initial research interests and the emerging data. We do not force preconceived ideas and theories directly upon our data. However, what might stand as a viable means of gathering data to one grounded theorist might be defined as forcing the data into a preconceived framework to another. For example, Glaser (1998: 94) cautions against preconceiving 'interview guides, units for data collection, samples, received codes, following diagrams, rules for proper memoing and so forth'.
Grounded theorists do agree on starting analysis with the data. We follow leads that we define in the data. Thus, I started with research interests m time and self-concept but also pursued other topics that my respondents defined as crucial. For example, 1 felt compelled to explore their concerns about disclosing illness, which I had not anticipated studying but which had emerged as a significant theme in the data. I studied how, when, and why ill people talk about their conditions. However, my interest in time alerted me to see whether people's accounts of disclosing their conditions changed over lime.
What kind of data should you gather for grounded theory studies? To ihe extent possible, I advocate going inside the studied phenomenon and gathering extensive, rich data about it, while simultaneously using grounded theory strategies to direct my data collection. Rich data reveal participants' thoughts, feelings, intentions, and actions as well as context and structure. My call for rich, detailed data means seeking full or 'thick' description (Geeríz, J973) such as writing extensive field notes of observations, collecting respondents' written personal accounts, and compiling detailed narratives of experience (such as transcribed tapes of interviews). Seidman (1998) advocates sequential intensive interviewing to build trust and to elicit detailed data. Transcribed tape recordings of interviews provide details for nuanced views and reviews of data, f find that studying the transcriptions gives me new insights and more codes with which to work, in contrast, Glaser (1998) argues that transcribing wastes lime and fosters becoming lost in data.
Grounded theorists take different, sometimes contradictory approaches to data collection, although all assume that the strength of grounded theory lies in its empirical foundation. Glaser (1992; 1998) consistently stresses discovering what is happening in the setting without forcing the data into preconceived categories through such errors as applying extant theories to it, assuming the significance of demographic variables (such as age, sex, race, marital status, and occupation; also called face-sheet variables) before beginning the study, imposing evidentiary rules (a priori prescriptions about
88       Qualitative Psychology
whal stands as sufficient evidence) on the data, or failing to make theoretical distinctions with empirical description. However, he also advocates short cuts such as moving quickly from one empirical world to another to develop a category, not transcribing interviews, and accepting a group's overt statements about itself, a practice which may obfuscate members' fundamental concerns. Such short cuts can cause problems. Researchers may obtain only a surface view of a group when they move quickly from one research site to another. In addition, people may not offer much beyond a public relations viewpoint until they trust the researcher. Furthermore, members may reveal their most important values and priorities through actions and assumptions, not careful statements. In effect, short cuts may curtail discoveries, miss basic social processes, overlook subtle meanings, and force data into categories prematurely. Strauss and Oorbin (1990) imply that concrete observed behaviour with scant interpretation constitutes solid data when they give hypothetical examples that lack nuanced description of a setting and the social actors and interaction within it. Their hypothetical example of the restaurant hostess describes overt movements and, thus, misses what actual hostesses do, feel, and think about their situations (1990: 63-5). However, their interview data in Unending Work and Care (Corbin and Strauss, 1.988) contains research participants' nuanced stories and meaningful statements.
Situate your data within settings and scenes, collective meanings, and individual interpretations, actions, and processes. Then, your descriptions will have more substance and form than mere observed behaviour. Even if you have detailed raw data, such as the typed transcription of a patient conference, elaborate on it. Provide the context by describing the structure of the conference, the events preceding it, the players in it, and their unstated concerns (if known or implicit in their non-verbai behaviour). Similarly, place a personal interview into perspective by describing the situation, the interaction, the person's affect, and your perception of how the interview went. Thorough written texts give you data to study. In short, get as much material down on paper as possible.
Rich data afford views of human experience that etiquette, social conventions, and inaccessibility hide or minimize in ordinary discourse. To obtain rich data:
•     describe participants' views and actions in detail
•     record observations that reveal participants' unstated intentions
•     construct interview questions that allow participants to reflect anew on the research topic
*     look for and explore taken-for-granted meanings and actions.
Grounded theory      89
Tell me about', 'how', 'what', and 'when' questions yield rich data, particularly when you buttress them with queries to elaborate or to specify, such as 'Could you describe -— further' (for a sample interview guide, see Charmaz, 2001). Look for the 'urns' and 'you know's'; explore what they Indicate. How might they reflect a struggle to find words? When might a 'you know' signal taken-for-granted meanings? What do long pauses indicate? When might 'you know' seek the interviewer's concurrence or suggest that the respondent is struggling to articulate an experience? In my research, however, respondents' stories about illness often spilled out non-stop. For example, Christine Danforth stated:
if you have lupus, I mean one day it's my liver; one day it's my joints; one day it's my head, and it's like people really think you're a hypochondriac if you keep complaining about different ailments. . . . it's like yon don't want to say anything because people <irc going to start thinking, you know, 'God, don't go near her, all she is - is complaining about this.' And I think that's why I never say anything because I feel like everything I have is related one way or another to the lupus but mos! of the people don't know ľ have lupus, and even those that do are not going to believe that ten different ailments are the same thing. And 1 don't want anybody saying, you know, they don't want to come around me because 1 complain. (Charmaz, 1991a: 114-15)
Through obtaining rich data, researchers gain thorough knowledge of liieir studied empirical worlds or research problems, Glaser (1998) advocates moving across different empirical settings quickly to seek comparative data about an emerging category. Instead, J recommend that novices gain a ihorough empirical grounding in one arena before exploring another. Then you may discern what participants mean and how they define their experiences, and subsequently interpret these data starting from your respondents' points of view. You may see other things in the data as well, because you bring different perspectives and concerns to it than do your participants, liiere I adopt the positivist assumption that researchers have the responsibility to find what is 'there' and that we can act on this responsibility because we already share language and meanings with those we study, or we í ;m learn them.) Rich data allow you to trace events, delineate processes, and make comparisons.
Throughout a grounded theory research project, you increasingly focus your data collection because your analytic work guides you to gather more •specific data. The grounded theorist's simultaneous involvement, in data gathering and analysis is explicitly aimed toward developing theory. Grounded theory ethnographers, for example, move from attempting to capture the whole round of life to focused areas to explore, observe, and analyse. Grounded
Qualitative Psychology
theory interviewers adapt their initial interview guides; they add areas id explore and delete extraneous questions.
Grounded theorists follow leads to develop their emerging theoretical categories (Glaser, 1978). Other qualitative researchers may produce thick description of concrete behaviour without filling out, extending, or refining theoretical concepts or making theoretical connections. In contrast, grounded theorists use thick description to ask theoretical questions. For example, young adults agonized over telling room-mates, acquaintances, and dates about their conditions. Their stones sparked my interest in dilemmas of disclosing illness. Rather than obtaining thick description only about their difficulties in disclosing, 1 began to ask analytic questions about disclosing as a process and then gathered data that illuminated that process. These questions included:
«    What are the properties of disclosing?
«    Which social psychological conditions foster disclosing? Which inhibit it?
»    How does disclosing compare with other forms of telling?
a    How, if at all, does disclosing change after the person becomes accustomed to his or her diagnosis?
•    What strategies, if any, do people use to disclose? When do they use them?
Researchers may adopt several grounded theory strategies to gather descriptive accounts without following the analytic steps that make their work theoretical. Listen closely to your respondents; attempt to learn unstated and assumed meanings of their statements; and shape your emerging research questions to obtain data that illuminate your theoretical categories. Then you will be doing grounded theory.
Defining Meanings and Processes
The grounded theory emphasis on studying processes moves research away from static analyses. We emphasize what people are doing, an emphasis which also leads to understanding multiple layers of meanings of their actions. These layers could include a person's 1) stated explanation of his or her action, 2} unstated assumptions about it, 3) intentions for engaging in it, 4) effects on others, and 5} consequences for further individual action and interpersonal relations. Throughout the research process, look at action in
Grounded theory      91
(elation to meaning to help you obtain thick description and develop your categories. How do you study meaning?
Some grounded theorists believe they can readily discover what is ■tijitiilicant in the research setting. Glaser (1992) states that the significant l.wues are apparent in the field selling and, therefore, so are the significant data. He warns that anything else preconceives the ensuing research. 1 differ on these points. The most important issues to study may be hidden, tacit, or elusive. We probably struggle to grasp them. The data we 'find' and the meanings we attribute to them reflect this struggle. Neither data nor meaningful interpretation of them simply await the researcher. Unlike Giaser, 1 assume that we are part of the meanings that we observe and define. In short, our understanding of respondents' meanings emerges from a particular viewpoint and the vocabulary that we invoke to make sense of them.
A researcher has topics to pursue; research participants have goals, thoughts, feelings, and actions. Your research questions and mode of inquiry shape your subsequent data and analysis. Thus, you must become seif-aware of why and how you gather data. You learn to sense when you are gathering i ich, useful data that do not undermine or demean your respondent(s). Not surprisingly, then, 1 believe the grounded theory method works best when the grounded theorist engages in data collection as well as data analysis phases of research. This way, you can explore nuances of meaning and process that hired hands might easily miss.
Respondents' stories may tumble out or the major process in which people are engaged may jump out at you. Sometimes, however, respondents may not be so forthcoming nor may major processes be so obvious. Even if ihey are, it may take more work to discover the subtlety and complexity of respondents' intentions and actions. The researcher may have entered the implicit worid of meaning, but not of explicit words. For example, many of my participants spoke of incidents in which their sense of social and persona! worth was undermined. They complained, told stories, and expressed incredulity. 1 began to see their accounts as stories of suffering {Charmaz, 1999). These stories reflected more than a stigmatized identity - but what? Gradually, I pieced their stories together in a hierarchy of moral status that catapults downward as health fails, resources wane, and difference increases. Sufferers talked about loss, not moral status. Yet everything they said relied on assumptions of moral status.
The further we go into implicit meanings, the more we may conceptualize them with abstract ideas that encapsulate the experiences that gave rise to these meanings. For example, when exploring implicit meanings of 'bad days', I defined them according to my participants' evaluations of intensified intrusiveness of illness; reduced control over mind, body, and actions; and curtailed choices and actions. I synthesized, condensed, and conceptualized participants' statements to make their tacit understandings explicit. Thus, we speak in our categories at this point, rather than reproduce
92      Qualitative Psychology
Grounded theory       93
participants' words. Some meanings are so well understood thai they remain unsialed and assumed; others remain felt, hut pariicipants have no words to voice them.
For certain topics, close study and direct questioning may suffice. For other topics, you may need to redirect inquiry. Because our language contains few words with which to talk about time, many of my research participants' altitudes toward and actions concerning time remained unspoken and taken for granted. Yet their stories about illness often depended on conceptions of time and referred to implicit qualities of experienced time. For example, Christine Danforth's statement above referred to the quality and unevenness of her days. If a researcher plans to explore such areas, then he or she often needs to devise ways to make relevant observations or to construct questions that will foster pertinent responses. To illustrate, 1 asked my respondents questions such as, 'As you look back on your illness, which events stand out in your mind?' and 'What is a typical weekday like for you?' Glaser (1992) might say 1 force the data here by asking preconceived questions of it. Instead, I generate data by investigating taken-for-granted aspects of life. At whatever level you attend to your participants' meanings, intentions, and actions, you can create a coherent analysis by using grounded theory methods. Hence, the method is useful for fact-finding descriptive studies as well as more conceptually developed theoretical statements.
Perhaps the most important basic rule for a grounded theorist is: study your emerging data (Glaser, 1978; 1992). Studying the data sparks your awareness of respondents' implicit meanings and taken-for-granted concerns. How do you study data? From the very start, transcribe your audiotapes yourself or write your own field notes rather than, say, dictating them to someone else. Studying your data prompts you to learn nuances of your research participants' language and meanings. Subsequently, you learn to define the directions where your data can take you. Through studying interview audiotapes, for example, you attend closely to your respondents' feelings and views. They will live in your mind as you listen carefully over and over to what they were saying. For example, one student in my class remarked:
What an impact the words had on me when 1 sat home alone transcribing the tapes. I was more able to hear and feel what these women were saying to me. 1 realized how, at times, 1 was preoccupied with thoughts of what my next question was, how my eye contact was, or hoping we were speaking loud enough for the tape-recorder. (Charmaz, 1991b: 393)
If you attend to respondents' language, you can adapt your questions to fit their experiences. Then you can learn about their meanings rather than make assumptions about what they mean. For example, when my
respondents with chronic illnesses often talked about having 'good days' and 'bad days', 1 probed further and asked more questions around my respondents' taken-for-granted meanings of good and bad day;,, i asked questions such as: 'What does a good day mean to you?', 'Could you describe what a bad day is?', 'What kinds of things do you do on a good day?', and 'How do these activities compare with those on a bad day?' By comparing interview accounts, 1 discovered that good days meant that pariicipants' temporal and spatial horizons expanded and that possibilities increased for realizing the selves they wished to be. But had I not followed tip and asked respondents about the meanings of these terms, their specif c properties would have remained implicit.
Certainly, starling the research with strong data-gathering skills helps. A skilled researcher knows when to ask more questions or to make more focused observations. Nevertheless, novices can make remarkable gains in skill during a brief time by attending closely to their methods and by studying their data. By gathering rich data and by making meanings explicit, you will have solid material with which to create your analysis.
Coding the Data
Coding is the process of defining what the data are about. Unlike quantitative data, in which preconceived categories or codes are applied to the data, grounded theorists create their codes by defining what they see in the data. (lodes emerge as you scrutinize your data and define meanings within them. This active coding forces you to interact with your data again and ask questions of them. (Thus, the interactive nature of grounded theory research is not limited to data collection, but also proceeds throughout the analytic work.) As a result, coding may lake you into unforeseen areas and new research questions.
Coding is the pivotal link between collecting data and developing an emergent theory lo explain these data, ft consists of at least two phases: an initial phase involving the naming of each line of data followed by a incused, selective phase that uses the most significant or frequent initial codes to sort, synthesize, and organize large amounts of data.
While coding, you use 'constant comparative methods' (Glaser and Strauss, 1967) to establish analytic distinctions - and thus make comparisons at each level of analytic work. At first, you compare data with data to lind similarities and differences. For example, compare interview statements within the same interview and compare statements in different interviews. When conducting observations of an activity, compare what happens on one day with the same activity on subsequent days. Next, you can ask, 'What category or property of a category does this incident indicate?' (Glaser, 1992: 39). initial coding entails examining each line of data
Qualitative Psychology
Grounded theory      95
and defining the actions or events iliat you see as occurring in it or as represented by it - line-by-line coding (see Box 5.1). Compare incident with incident; then, as your ideas take hold, compare incidents to your eonap totalization of incidents coded earlier. That way you can identify properties of your emerging concept.
Line-by-line coding means naming each line on each page of youi written data (Glaser, 1978), although you may not always have complete sentences. Through line-by-line coding, you take an analytic stance toward your work and, simultaneously, keep close to your data. Coding leads directly to developing theoretical categories, some of which you may defint in your initial codes. You build your analysis from the ground up without taking off on theoretical flights of fancy. Line-by-line coding also helps you to refrain from imputing your motives, fears, or unresolved personal issues to your respondents and to your collected data. Some years ago, a young man in my undergraduate seminar conducted research on adaptation to disability. He had become paraplegic himself when he was hit by a car wh:k bicycling. His ten in-depth interviews were filled with stories of courage, hope, and innovation. His analysis of them was a narrative of grief, angei, and loss. When 1 noted that his analysis did not reflect his collected material, he realized how his feelings coloured his perceptions of other people's disabilities. His was an important realization. However, had he assiduously done line-by-iine coding, he might have arrived at it before he handed in his paper.
From the standpoint of grounded theory, each idea should earn its way into your analysis (Glaser, 1978). if you apply theoretical concepts from your discipline, you must ensure that these concepts work. Do they help you understand what the data indicate? Can you explicate what is happening in this line of data? If they do not, use other terms that do.
Line-by-line coding forces you to think about the material in new ways that may differ from your research participants' interpretations. For Thomas (1993), a researcher must take the familiar, routine, and mundane and make it unfamiliar and new. Line-by-line coding helps you to see the familiar in a new light. You also gain distance from both your own and your participants' taken-for-granted assumptions about the material, so that you can see it in new light.
If your codes define another view of a process, action, or belief than your respondents hold, note that. Your task is to make analytic sense of the material. How do you make analytic sense of the rieh stories and descriptions you are compiling? First, look for and identify what you see happening in the data. Some basic questions may help:
o    What is going on?
*     What is the person saying?
»    What do these actions and statements lake for granted?
o     How do structure and context serve to support, maintain, impede, or change these actions and statements?
Try to frame your codes in as specific terms as possible - and keep them short. Make them active. Short, specific, active codes help you define processes in the data that otherwise may remain implicit. What you see in these tlata relies in part upon your prior perspectives. Rather than seeing your perspectives as truth, try to see them as representing one view among many. That way, you may gain more awareness of the concepts that you employ. For example, try not to assume that respondents repress or deny significant 'facts' about their lives, instead, look for how they understand their situations before you judge their attitudes and actions through your own assumptions. Seeing the world through their eyes and understanding the logic of their experience brings you fresh insights. Afterwards, if you still invoke previously held perspectives as codes, you will use them more consciously rather than automatically.
In the example in Box 5.1 of line-by-line coding, my interest in time jnd self-concept comes through in the first two codes. Note how I kept the codes active and close to the data.
Initial codes often range widely across a variety of topics. Because even a short statement or excerpt may address several points, it could illustrate several different categories. 1 could use the excerpt in Box 5.1 to show how avoiding disclosure serves to control identity. I could also use it to show either how a respondent learns that other people see his or her illness as inexplicable or how each day is unpredictable. Having multiple interviews allows me to see how social and emotional isolation begins and progresses.
Initial codes help you to separate data into categories and to see processes. Line-by-line coding frees you from 'going native', that is, becoming so immersed in your respondents' world-view that you accept it without question. Then you fail to look at your data critically and analytically. Being critical about your data does not necessarily mean being critical of your research participants. Instead, being critical forces asking yourself questions about your data. These questions help you to see actions and to identify the significant processes. Such questions include:
•     What process is at issue here? How can I define it?
a    What are people doing?
•    Under which conditions does this process develop?
96      Qualitative Psychology
Box 5.1   Initial coding: line-by-line coding
Shifting symptoms, having
inconsistent days lnterpreting images ol sell
given by others Avoiding disclosure Predicting rejection Keeping others unaware Seeing symptoms as
connected Having others unaware Anticipating disbelief Controlling others' views Avoiding stigma Assessing potential tosses
and risks ot disclosing
Meaning ot the CVA Feeling to reed to live one
day at a time Having a worried past Earlier losses Difficult living one day at
lime; concentrate on today Giving up future orientation Managing emotions thru
living one day at time reducing I if e-threatening risk
Excerpt 1, Chrisline Danforth, Age 37, lupus erythematosus, Sjogren's syndrome, back injuries. Lupus erythematosus is a systemic, inflammatory auto-immune disease of the connective lissue that affects vital organs as well as joints, muscles, and nerves. Sjogren's sydrome is a related autoimmune inflammatory disease characterized by dry mucous membranes of the eyes and mouth.
If you have lupus. I mean one day it's my liver; one day it's my joints; one day it's my head, and it's like people really think you're a hypochondriac if you keep complaining about different ailments. . . . It's like you don't want to say anything because people are going to start thinking, you Know, 'God, don't go near her, all she is -is complaining about this.' And I think that's why I never say anything because I feel like everything I have is related one way or another to the lupus but most of the people don't know I have lupus, and even those that do are not going to believe that ten different ailments are the same thing. And I don't want anybody saying, you know, fthat] they don't want to come around me because i complain.
Excerpt 2, Joyce Marshall, age 60, minor heart condition, recent small cerebral vascular accident (CVA) (stroke). In her case, the stroke left her with weakness, fatigue, and slowed responses when tired.
I have to see it [her CVA] as a warning. I can't let myself get so anxious. I have to live one day at a time. I've been so worried about John [her husband who had had life-threatening heart attacks and lost his job three years before retirement) and preparing to get a job [her first in 38 yearsj. . . . it's just so hard with all this stress ... to concentrate on what I can do today. I always used to look to the future. I can't now; it upsets me too much. I have to live one day at a time now or else there may not be any me.
Grounded theory      97
»    How does the research participants) think, feel, and act while involved in this process?
•    When, why, and how does the process change?
e    What are the consequences of the process?
Through coding each line of data, you gain insights about what kinds of data to collect next. Thus, you distil data and direct further inquiry early in the data collection. Line-by-line coding gives you leads to pursue. If, for example, you identify an important process in your fifteenth interview, you can return to earlier respondents and see whether that process explains events and experiences in their lives, ff not, you can seek new respondents who can illuminate this process. Hence, your data collection becomes more focused, as does your coding.
After you have established some strong analytic directions through your initiai iine-hy-line coding, you can begin focused coding to synthesize and explain larger segments of data. Focused coding means using the most significant and/or frequent earlier codes to sift through large amounts of dala. Thus, focused coding is more directed, selective, and conceptual than line-by-line coding (Glaser, 1978). Focused coding requires decisions about which initial codes make the most analytic sense and categorize your data most accurately and completely. Yet, moving to focused coding is not entirely a linear process. Some respondents or events make explicit what was implicit in earlier respondents' statements or prior events. An 'Aha! Now 1 understand' experience prompts you to study your earlier data afresh. Then you may return to earlier respondents and explore topics that had been glossed over, or that may have been too implicit or unstated to discern.
The strength of grounded theory coding derives from this concen-inited, active involvement in the process. You act upon rather than passively read your data. Through your actions, new threads for analysis become apparent. Events, interactions, and perspectives that you had not thought of before come into analytic purview. Focused coding checks your preconceptions about the topic.
In the first excerpt in liox 5.2, I selected the codes 'avoiding disclosure' and 'assessing potential losses and risks of disclosing' to capture, synthesize, ;md understand the main themes in the statement. In the second, the following codes were most useful: 'feeling forced to live one day at a tíme', 'concentrating on today', 'giving up future orientation', 'managing emotions', and 'reducing life-threatening risk'. Again, 1 tried to keep the codes active and close to the data. Through focused coding, you can move across interviews and observations and compare people's experiences, actions, and Interpretations. Note how the codes condense data and provide a handle to them.
Qualitative Psychology
Box 5.2   Focused coding
Avoiding disclosure
Assessing pole n Hal losses grid risks of disclosing
Feeling forced to live one day al a lime
Concentrating on today Giving up I u tu re orientation Managing emotions Reducing life-threatening risk
Excerpt 1, Christine Danforth, age 37, lupus erythematosus, Sjogren's syndrome, back injuries
If you have lupus, I mean one day it's my liver; one day it's my joints; one day it's my head, and it's like people really think you're a hypochondriac if you keep complaining about different ailments. . . . It's Sike you don't want to say anything because people are going to start thinking, you know, 'God, don't go near her, all she is -is complaining about this.' And i think that's why I never say anything because I feel like everything I have is related one way or another to the fupus but most of the people don't know I have lupus, and even those that do are not going to believe that ten different ailments are the same thing. And i don't want anybody saying, you know, [that] they don't want to come around me because I complain.
Excerpt 2, Joyce Marshall, age 60, minor heart condition, recent small CVA (stroke)
! have to see it [her CVA] as a warning, i can't let myself get so anxious. I have to live one day at a time. I've been so worried about John [her husband who had had life-threatening heart attacks and lost his job thtee years before retirement] and preparing to get a job [her first in 38 years]. . . . it's just so hard with all this stress ... to concentrate on what I can do today. I always used to look to the future. I can't now; it upsets me too much, i have to live one day at a time now or else there may not be any me.
Strauss and Corbin (1990) also introduce a third type or coding, axial coding, to specify the dimensions of a category. The purpose is to sort, synthesize, and organize large amounts of data and reassemble them in new ways after open coding (Cresswell, 1998). For example, when studying disclosure of illness, i re-examined the data 1 had coded during open coding. Then 1 coded for the range between spontaneous statements and staged pronouncements. When engaged in axial coding, the researcher also links categories with sub-categories, and asks how they are related. Hence, i linked forms of teliing explicitly to the relative absence or presence of strategizing.
Grounded theory      99
Alter discovering thai people invoked different forms o! telling, 1 looked more closely at the context of their telling and the conditions affecting how und whom they told, as well as their stated intentions for telling. 1 coded for how, when, and why they changed their earlier forms of telling. These strategies may lead to charting causes and conditions of the observed phenomenon. Whether axial coding helps or hinders remains a question. Whether it differs from careful comparisons also is questionable. At best, it helps to clarify; at worst, it casts a technological overlay on the data. Although intended to obtain a more complete grasp of the studied phenomena, axial coding may make grounded theory cumbersome (Robrecht, 1995).
Raising Focused Codes to Conceptual Categories
focused coding moves your analysis forward in two crucial steps: 1) to establish the content and form of your nascent analysis and, 2) to evaluate and clarify your categories and the relationships between them. First, assess which codes best capture what you see happening in your data. Raise them to conceptual categories for your developing analytic framework - give them conceptual definition and analytic treatment in narrative form. Thus, you go beyond using a code as a descriptive tool to view and synthesize data.
Categories explicate ideas, events, or processes in your data - and do so in telling words. A category may subsume common themes and patterns in several codes. For example, my category of 'keeping illness contained' included 'packaging illness', that is, treating it 'as if it were controlled, delimited, and confined to specific realms, such as private life', and 'passing', which means 'concealing illness, maintaining a conventional self-presentation, and performing like unimpaired peers' (Charmaz, 1991a: <>6~8). Again, make your categories as conceptual as possible - with abstract power, genera! reach, analytic direction, and precise wording. Simultaneously, remain consistent with your data. By making focused codes active (to reflect what people are doing or what is happening) and brief, you can view them as potential categories. Processes gain visibility when you keep codes active. Succinct focused codes lead to sharp, clear categories. That way, you can establish criteria for your categories to make further comparisons.
Grounded theorists look for substantive processes that they develop from their codes. 'Keeping illness contained', 'packaging illness', and 'living one day at a time' above are three such processes. As grounded theorists create conceptual handles to explain what is happening in the setting, they may move toward defining generic processes (Prus, 1987). A generic process cuts across different empirical settings and problems; it can be applied to varied substantive areas. The two codes above, 'avoiding disclosure' and 'assessing potential losses and risks of disclosing', reflect fundamental, generic processes of personal information control. Although these processes
Qualitative Psychology
describe choices people wilh illness make in disclosing information, people with other problems may treat information control similarly. For sociologists, generic processes are basic to social life; for psychologists, generic-processes are fundamental for psychological existence. Thus, a grounded theorist can elaborate and refine the generic process by gathering more data from diverse arenas where this process is evident. For example, personal information control and choices in disclosing are often problematic for homosexuals, sexual abuse survivors, drug users, and ex-convicts, as well as for people with chronic conditions. Concentrate on analysing a generic process that you define in your codes; then you can raise codes relevant to theoretical categories that lead to expianations of the process and predictions concerning it. These categories reflect what you think about the data as well as what you find in them. As Dey (1999) observes, categorization in grounded theory is more complex and problematic than its originators suggest and involves making inferences as well as classifications.
As you raise a code to a category, you begin to write narrative statements in memos, as I outline beiow, that:
•     explicate the properties of the category
•     specify the conditions under which the category arises, is maintained, and changes
•     describe its consequences
•     show how this category relates to other categories.
Categories may consist of in vivo codes that you take directly from your respondents' discourse, or they may represent your theoretical or substantive definition of what is happening in the data. For example, my terms 'good days and bad days' and 'living one day at a time' came directly from my respondents' voices. In contrast, my categories 'recapturing the past' and 'time in immersion and immersion in time' reflect theoretical definitions of actions and events. Furthermore, categories such as 'pulling in', 'facing dependency', and 'making trade-offs' address my respondents' substantive realities of grappling with a serious illness. í created these codes and used them as categories, but they reflect my respondents' concerns and actions. Novice researchers may find that they rely most on in vivo and substantive codes. What results is often a grounded description more than a theory. Nonetheless, studying how these codes fit together in categories can help you treat them more theoretically.
Through focused coding, you build and clarify your category by examining all the data it covers and by identifying variations within it and between other categories. You also wi!I become aware of gaps in your
Grounded theory    1
Oox 5.3   The category of 'existing from day to day'
Existing from day to day occurs when a person plummets into continued crises that rip life apart. I! reflects a toss of control of health and the wherewithal to keep life together.
Existing from day to day means a constant struggle for daily survival. Poverty and lack of support contribute to and complicate that struggle. Hence, poor and isolated people usually plummet further and faster than affiueni individuals with concerned families. Loss of control extends to being unable to obtain necessities -food, shelter, heat, and medical care.
The struggle to exist keeps people in the present, especially if they have continued problems in getting fhe basic necessities that middle-class adults take for granted. Yet other problems can assume much greater significance for these people than their iliness - a violent husband, a runaway child, an alcoholic spouse, or the overdue rent.
Living one day at a time differs from existing from day to day. Living one day at a time provides a strategy for controlling emotions, managing life, dimming the future, and getting through a troublesome period. St involves managing stress, illness, or regimen, and dealing with these things each day to control them as best one can. it means concentrating on the here and now and relinquishing other goals, pursuits, and obligations (Channaz, 1991a: 185).
analysis. For example, I developed my category of 'existing from day to day' when ! realized that 'living one day at a time' did not fully covet impoverished people's level of desperation, in short, I had data about a daily struggle to survive that were not subsumed by my first category of living one day at a time. The finished narrative can be seen in Box 5.3.
Note the comparisons between the two categories above. To generate categories through focused coding, you need to compare data, incidents, contexts, and concepts. Making the following comparisons helps:
o    comparing different  people (about  their beliefs,  situations,  actions, accounts, or experiences)
*    comparing data from the same individuals at different points in time
e    comparing specific data with the criteria for the category
e    comparing categories in the analysis with other categories.
As ! compared different people's experiences, I realized that some peopäe's situations forced them into the present. 1 then looked at how my rendering of living one day at a time did not apply to them. 1 reviewed
102    Qualitative Psychology
earlier interviews and began 10 seek published accounts lhal might clarify the comparison. As is evident in the distinctions between these |wo categories above, focused coding prompts you to begin to see the relationships and patterns between categories.
Memo-writing
You may have thought of memos as business communications to state policies, procedures, and proposals. However, in grounded theory, memos serve analytic purposes. Memo-writing consists of taking your categories apart by breaking them into their components. It is the pivotal intermediate step between defining categories and the first draft of your completed analysis. This step spurs you to develop your ideas in narrative form and fullness early in the analytic process. Memo-writing is the logical next step after you define categories; however, it is also helpful for clarification and direction during coding. Writing memos prompts you to elaborate processes, assumptions, and actions covered by your codes or categories. Memos also help you to identify which codes to treat as analytic categories, if you have not already defined them. (Then you further develop your category through more memo-writing.)
Think about including as many of the following points in your memos as is possible and useful:
•     defining each code or category by its analytic properties
•     spelling out and detailing processes subsumed by the codes or catgories
•     making comparisons between data and between codes and categories
«    bringing raw data into the memo
o    providing sufficient empirical evidence to support your definitions of the category and analytic claims about it
»    offering conjectures to check in the empirical research
•     identifying gaps in the analysis.
Grounded theorists look for patterns, even when focusing on a single case (Strauss and Glaser, 1970). Because they stress identifying patterns, grounded theorists typically invoke respondents' stories to illustrate points -rather than provide complete portrayals of their lives. By bringing raw data right into your memo, you preserve telling evidence for your ideas from
Grounded theory
lhe start of your analytic narratives. Through providing ample verbatim material, you not only ground the abstract analysis, but also lay the foundation for making claims about it. Including verbatim material from different sources permits you to make precise comparisons. Thus, memo-writing moves your work beyond individual cases through defining patterns.
Begin your memo with careful definitions of each category. That means you identify its properties or characteristics, look for its underlying assumptions, and show how and when the category develops and changes. To illustrate, 1 found thai people frequently referred to living one day at a time when they suffered a medical crisis or faced continued uncertainty. So 1 began to ask questions about what living one day at a time was like for them. From their responses as well as from published autobiographical accounts, 1 began to define the category and its characteristics. The term 'living one day at a time' condenses a whole series of implicit meanings and assumptions, it becomes a strategy for handling unruly feelings, for exerting some control over a now uncontrollable life, for facing uncertainty, and for handling a conceivably foreshortened future. Memo-writing spurs you to dig into implicit, unstated, and condensed meanings.
Start writing memos as soon as you have some interesting ideas and categories to pursue, if at a loss about what to write, elaborate on codes that you adopted repeatedly. Keep collecting data, keep coding, and keep refining your ideas through writing more and further developed memos. Some researchers who use grounded theory methods discover a few interesting findings early in their data collection and then truncate their research. Their work lacks the 'intimate familiarity' with the setting or experience that Lofiand and Lofland (1995) avow meets the standards for good qualitative research. Cover your topic in depth by exploring sufficient cases and by elaborating your categories fully.
Memo-writing frees you to explore your ideas about your categories. Treat memos as partial, preliminary, and eminently correctable, just note where you are on firm ground and where you are making conjectures. Then go back to the field to check your conjectures. Memo-writing is much like free-writing or prewriting (Elbow, 1981). Use memos to help you think about the data and to discover your ideas about them. You can write memos for your eyes only. Do not worry about verb tense, overuse of prepositional phrases, or lengthy sentences at this point. Just get your ideas down as quickly and clearly as you can. You are writing to render the data, not to communicate it to an audience. Later, after you turn your memo into a section of a paper, revise the material to make it accessible to a reader. Writing memos quickly without editing them fosters developing and preserving your natural voice. Then your memo reads as though written by a living, thinking, feeling human being rather than a pedantic social scientist. You can write memos at different levels of abstraction - from the concrete to the highly theoretical. Some of your memos will find their way directly into