104 Qualitative Psychology your first draft of your analysis. Set aside others wilh a different focus and develop them later. Direct much of your memo-writing to making comparisons, what Glaserand Strauss (1967: 105) call 'constant comparative methods'. This approach emphasizes comparing incidents indicated by each category, integrating categories by delineating their relationships, delimiting the scope and range of the emerging theory, and writing the theory. Hence, you compare one respondent's beliefs, stance, and actions with another respondent's, or one experience with another. If you have longitudinal data, compare a participant's response, experience, or situation at one point in tíme with that at another time. Then, as you become more analytic, start to make detailed comparisons between categories and then frame them into a theoretical statement. Through memo-writing, you distinguish between major and minor categories. Thus, you direct the shape and form of your emergent analysis. At each more analytic and abstract level of memo-writing, bring your data right into your analysis. Show how you build your analysis on your data in each memo. Bringing your data into successive levels of memo-writing ultimately saves time; you do not have to dig through stacks of material to illustrate your points. A section of a memo is provided in Box 5.4. Note that I first defined the category, 'living one day at a time', and pointed out its main properties. Then i developed aspects of living one day at a time such as its relationship to time perspective, which is mentioned here, and to managing emotions. The memo also covered how people lived one day at a time, the problems it posed as well as those it solved, and the consequences of doing so. Theoretical Sampling Memo-making leads directly to theoretical sampling, that is, collecting more data to illuminate your theoretical categories. Here, you sample for the purpose of developing your emerging theory, not for representation of a population or increasing the generali zabi lity of your results. Thus, you seek more cases or ask earlier participants about experiences that you may not have covered before. You need more data to be sure that your category accurately describes the underlying quality of your respondents' experiences. In contrast, quantitative researchers need to have random samples whose characteristics are representative of the population under study. Whereas survey researchers want to use sample data to make statistical inferences about the target population, grounded theorists are interested primarily in the fit between their data and the emerging theory. When I was trying to figure out how people with chronic illnesses defined the passage of time, í went back to several participants whom I had Grounded theory 105 Box 5.4 Example of memo-writing Living One Day at a Time Living one day at a time means dealing wilh illness on a day-to-day basis, holding future plans and even ordinary activities in abeyance while the person and, often, others deal with illness. When living one day at a time, the person feels that his or her future remains unsettled, that he or she cannot foresee the future or whether there will be a future. Living one day at a time allows the person to focus on illness, treatment, and regimen without becoming entirely immobilized by fear or future implications. By concentrating on the present, the person can avoid or minimize thinking about death and the possibility of dying. Relation to Time Perspective The felt need to live one day at a time often drastically alters a person's time perspective. Living one day at a time pulls the person into the present and pushes back past futures (the futures the person projected before illness or before this round of illness) so that they recede without mourning [their loss]. These past futures can slip away, perhaps almost unnoticed. [I then compare three respondents' situations, statements, and time perspectives.] interviewed before and asked them more focused questions about how they perceived times of earlier crisis and when time seemed to slow, quicken, drift, or drag. Because such topics resonated with their experiences, they even responded to esoteric questions. For example, when 1 studied their stories, i realized that chronically ill adults implicitly located their self-concepts in the past, present, or future. These time frames reflected the form and content of self and mirrored hopes and dreams for self as well as beliefs and understandings about self. Hence, I made 'the self in time' a major category. Thereafter, I explicitly asked more people whether they saw themselves in the past, present, or future. An elderly working-class woman said without hesitation: I see myself in the future now. If you'd asked where 1 saw myself eight months ago, I would have said, 'the past'. I was so angry then because I had been so active. And to go downhill as fast as I did - I felt life had been awfully cruel to me. Now I see myself in the future because there's something the Lord wants me to do. Here i sit all crumpled in this chair not being able to do anything for myself and still there's a purpose for me to be here- [Laughs.] í wonder what it could be. (Charmaz, 1991a: 256) 106 Qualitative Psychology Through theoretical sampling you can elaborate the meaning of your categories, discover variation within them, and define gaps between categories. Theoretical sampling relies on comparative methods for discovering these gaps and finding ways to (ill them. By the time you conduct theoretical sampling, you already have a set of explanatory categories that you can treat as concepts, in turn, these concepts are useful to understand many incidents or issues in your data (.Strauss and Corbin, 1990). 1 recommend conducting theoretical sampling later in the research to ensure that you have already defined relevant issues and allowed significant data to emerge. Otherwise, early theoretical sampling may bring premature closure to your analysis. Variation within the studied process will probably become apparent as you engage in theoretical sampling. It requires being selective about which data you seek and from whom you seek it. You may focus on certain experiences, events, or issues, not on individuals per se to develop your theoretical categories and to define how and when they vary. However, you are likely to gain more knowledge about the experiences, events, or issues that you seek to treat theoretically through observing or talking with individuals, l-'or example, one of my main categories was 'immersion in illness' (Charmaz, 1991a). Major properties of immersion include recasting life around illness, slipping into illness routines, pulling into one's inner circle, facing dependency, and experiencing an altered (slowed) time perspective. However, not everyone's time perspective changed. How could 1 account for that? By going back through my data, 1 gained some leads. Then 1 talked with more people about specific experiences and events that influenced their time perspective. Theoretical sampling helped me to refine the analysis and make it more complex. I then added a category, 'variations in immersion', to highlight and account for different experiences of immersion in illness. 1 filled out this category through theoretical sampling because I sensed variation earlier when comparing the experiences of people with different illnesses, different life situations, and different ages but had not made clear how immersion in illness varied and affected how these people experienced time. Subsequently, for example, 1 sampled to learn how illness and time differed for people who spent months in darkened rooms and how both varied when people anticipated later improvement or faced continued uncertainty. Thus, initial demographic variations in immersion led to useful theoretical understandings of variations in immersion itself. Making comparisons explicit through successive memos enabled me to draw connections that 1 did not initially discern. The memo became a short section of a chapter that begins as in Box 5.5 and then goes on to detail each remaining point. Theoretical sampling focuses further data collection to refine key categories in your research. You can then define them explicitly and identify their properties and parameters. Your subsequent memo-writing becomes more precise, analytic, and incisive. Theoretical sampling keeps you moving Grounded theory 107 Box 5.5 Variations in immersion A lengthy immersion in illness shapes daily lile and affects how one experiences time. Conversely, ways of experiencing time dialecticaliy affect the qualities of immersion in illness. The picture above of immersion and time has sharp outlines. What sources of variation soften or alter the picture of immersion and time? The picture may vary according to the person's 1} type of illness, 2) kind of medication, 3) earlier time perspective, 4) life situation, and 5) goals. The type of illness shapes the experience and way of relating to time. Clearly, trying to manage diabetes necessitates gaining a heightened awareness of timing the daily routines. But the effects of the illness may remain much more subtle. People with Sjogren's syndrome, for example, may have periods of confusion when they feel wholly out of synchrony with the world around them. For ihem, things happen too quickly, precisely when their bodies and minds function too slowly. Subsequently, they may retreat into routines to protect themselves. Lupus patients usually must retreat because they cannot tolerate the sun. Sara Shaw covered her windows with black blankets when she was extremely ill. Thus, her sense of chronological time became further distorted as day and night merged together into an endless flow of illness (Charmaz, 1991a: 93). L.........................------_-.......-.—____ , between targeted data collection and analytic memo-writing. You follow leads, check out hunches, and refine your ideas. Because theoretical sampling forces you to check ideas against direct empirical realities, you have solid materials and sound ideas with which to work. You gain confidence in your perceptions of your data and in your developing ideas about them. When do you stop gathering data? The standard answer is that you stop when your categories are 'saturated'. And categories are 'saturated' when new data no longer spark new insights. But researchers disagree about the meaning of saturation. As Janice Morse (1995) suggests, researchers proclaim saturation rather than prove that they have achieved it. Thus, like other qualitative approaches, the grounded theory approach shares the hazard of assuming that categories are saturated when they may not be. The kinds of research questions and the analytic level of the subsequent categories matter. Mundane questions may rapidly produce saturated but common categories, whereas novel questions may demand more complex categories and more sustained inquiry. Writing the First Draft After defining your theoretical categories fully, supporting them with evidence, and ordering them by sorting the memos you have written about 108 Qualitative Psychology Grounčed theory 109 them, you start writing the first draft of your paper. Writing is more than mere reporting. Instead, the analytic process proceeds through the writing of your report. Use your now developed categories to form sections of the paper. Show the relationships between these categories. When you have studied a process, your categories will reflect its phases. Yet you still need to make an argument for your reader as to why this process is significant. That means making your logic and purpose explicit. That may take a draft or two. Then outline your draft to identify your main points and to refine how you organize them. (But do not start your draft from an outline - use your memos.) As your argument becomes clearer, keep tightening it by reorganizing the sections of your paper around it. What place do raw data such as interview excerpts or field notes have in the body of your paper? Grounded theorists generally provide enough verbatim material to demonstrate the connection between the data and the analysis, but give more weight to the concepts derived from the data. To date, qualitative researchers do not agree on how much verbatim material is necessary. Compared to those qualitative studies that primarily synthesize description, grounded theory studies are substantially more analytic and conceptual. Our analytic focus encourages making theoretical relationships explicit and subordinating verbatim material to it (Glaser, 1978). Unlike some grounded theorists, 1 prefer to present many detailed interview quotations and examples in the body of my work. 1 do so to keep the human story in the forefront of the reader's mind and to make the conceptual analysis more accessible to a wider audience. After you have developed your conceptual analysis of the data, go to the literature in your field and compare how and where your work fits in with it - be specific. At this point, you must cover the literature thoroughly and weave it into your work explicitly. Then revise and rework your draft to make it a solid finished paper. Use the writing process to sharpen, clarify, and integrate your developing analysis. Through writing and rewriting, you can simultaneously make your analysis more abstract and your rendering and grounding of it more concrete. In short, you hone your abstract analysis to define essential properties, assumptions, relationships, and processes while providing sufficient actuai data to demonstrate how your analysis is grounded in people's experience. Conclusion The inductive nature of grounded theory methods assumes an open, flexible approach. Hence, you shape your methodological strategies while engaged in your research rather than having them planned before beginning the data collection. Similarly, you shape and alter the data collection to pursue the most interesting and relevant material. The purpose of grounded theory is to develop a theoretical analysis of the data that fits the data and has relevance to the area of study. The systematic procedures of grounded theory enable qualitative researchers to generate ideas. In turn, these ideas may later be verified through traditional quantitative methods. Nonetheless, as Glaser and Strauss originally claimed, grounded theory qualitative works stand on their own because they: 1) explicate basic (generic) processes in fhc data; 2) analyse a substantive field or problem; 3) make sense of human behaviour; 4) provide flexible, yet durable, analyses that other researchers can refine or update; and 5) have potential for greater generaiizability (for example, when conducted at multiple sites) than other qualitative works. But do most researchers who claim to do grounded theory research actually construct theory? No, not at this point. At present, most construct conceptual analyses of a particular experience instead of creating substantive or formal theory. These researchers pursue basic questions within the empirical world and try to understand the puzzles it presents. They emphasize analytic categories that synthesize and explicate processes in the worlds they study rather than tightly framed theories that generate hypotheses and make explicit predictions. Many researchers engage in grounded theory coding and memo-making but do not conduct theoretical sampling or pursue extensive analysis of their categories. However, grounded theory methods provide powerful tools for taking conceptual analyses into theory development. For this reason, grounded theory methods offer psychologists exciting possibilities for revisioning psychological theory as well as useful strategies for rethinking psychological research methods. Note 1 The interview was conducted by a trained student assistant for a mini-grant titled, 'Identity Hierarchies and Identity Goals: Adaptation to Loss Among the Chronically 11!', awarded by Sonoma State University. Ail names of interview participants have been changed. Acknowledgement This chapter is an earlier version of Charmaz (1995). f am indebted to Jennifer Dunn, Sachiko Kuwaura, and Jonathan Smith for comments on the first version of these ideas and to Judith Abbott, Emiliano Ayala, Lynn Cominsky, Carole Heathe, Jane Hood, Erich Lehmann, Catherine Nelson, Jonathan Smith, and anonymous reviewers for critiques of this chapter. Qualitative Psychology Further Reading Charma^, K. (2001) 'Grounded theory; methodobgy and theory construction', in N.j. Smeiser and P.8. Bakes (eds), International Encyclopedia of the Social and Behavioral Sciences. Amsterdam: Pergamon. pp. 6396-9. This article provides a succinct statement of the logic of theory construction in grounded theory. Charmaz, K, and Mitchell, R.C. (2001) 'Grounded theory in ethnography', in P. Atkinson, A. Coffey, S. Deiamont,J. Lofland and LH. Loffand (eds), Handbook of Ethnography. London: Sage. pp. 160-74. This chapter shows how grounded theory and traditional ethnography can complement each other. Glaser, B.C. (T9?8) Theoretical Sensitivity. Mill Valley, CA: Sociology Press. This book contains the most definitive statement of the original grounded theory method. Glaser, B.G. and Strauss, A.L. (1967) The Discovery of Grounded Theory. Chicago: Aldine. This book provides the first statement of the grounded theory method and presents the rationale for it and for qualitative research, more generally. Strauss, A.L. (1987) Qualitative Analysis for Social Scientists. New York: Cambridge University Press. This book provides a 'hands-on' description of how Anselm Strauss taught grounded theory to graduate students through group participation. Chapter 6 Narrative psychology Michael Murray Recently, the British writer A.S. Byatt published a collection of essays about narrative in which she argued that narrative lies at the heart of being human. Narration, she claimed, 'is as much part of human nature as breath and the circulation of the blood' (Byatt, 2000: 21}. Narrative pervades our everyday life. We are born into a narrative world, live our lives through narrative and afterwards are described in terms of narrative. Until recently, the study of narrative was considered as being of interest only to literary or foiklore critics (e.g., Brooks, 1985; Propp, 1968}, but it increasingly has assumed greater importance in the social sciences. Narrative is concerned with the human means of making sense of an ever-changing world. It is through narrative that we can bring a sense of order to the seeming disorder in our world, and it is though narrative that we can begin to define ourselves as having some sense of tempora! continuity and as being distinct from others. The aim of this chapter is to consider some of the theoretical issues around narrative psychology and some methodological issues around forms of narrative research. Narrative Psychology History of N'arraitve Psychology Recent interest in the study of narrative arose as part of the generai turn to language that occurred in the social sciences in the 1980s. Within psychology, three classic texts marked the specific narrative turn. The first was Narrative Psychology: The Storied Nature of Human Conduct, edited by Theodore Sarbin (1986). This collection amounted to a manifesto for the transformation of psychology. Sarbin contrasted the machine metaphor which he argued, underlay much of mainstream psychology with that of the narrative metaphor. He summarized the impiications of this alternative model: 111 112 Qualitative Psychology Narrative psychology 113 . In giving accounts of ourselves or of others, we are guided by I narrative plots. Whether for formal biographies or aulobiogra-i phies, for psychotherapy, for self-disclosure, or for entertainment, we do much more than catalog a series of events. Rather, we render the events into a story, (p. 23) In a later interview with Heaven (1999), Sarbin described how this idea arose in his discussion with theorists in the humanities. At first, he recalled, he did not distinguish between narrative as a mode of representation and narrative as an ontological form. However, over time, he became convinced that the latter stronger form of narrative was more appropriate. As he emphasized in his interview with Heaven (1999), stories have ontological status. We are aiways enveloped in stories. The narrative for human beings is analogous to the ocean for fishes, {p. 303) According to this argument, narratives are not just ways of seeing the world but we actively construct the world through narratives and we also live through the stories told by others and by ourselves - they have ontologica! status. The book edited by Sarbin (1986) also contains a chapter by Ken and Mary Gergen (1986) on the structure of narratives which was an extension of an earlier article (Gergen and Gergen, 1984), in which they argued that narratives are social constructions that are developed in everyday social interaction. They are a shared means of making sense of the world. They aiso have a certain structure. Gergen and Gergen identified three structures ' which they felt, organized many narratives; that is, the progressive, in which i there is movement towards a goal; the regressive, in which the reverse . occurs; and the stable, in which there is little change. Later in this chapter, we will consider the value of this model in the analysis of narrative accounts. The second important book was Acts of Meaning by Jerome flruner (1990), which followed his earlier Actual Minds: Possible Worlds (Bruner, 1986). In these books, Bruner argued that there are two forms of thinking: the paradigmatic and the narrative. The former is the method of science and is based upon classification and categorization. The alternative narrative approach organizes everyday interpretations of the world in storied form. The challenge of contemporary psychology is to understand this everyday form of thinking. Bruner identified a number of defining properties of narrative, including the following; 1. It is composed of a unique sequence of events, mental states and happenings involving human beings as characters or actors. 2. it can be 'real' or 'imaginary'. 3. If specializes in the forging of links between the exception;:! and the ordinary. These properties help us understand narrative as ways of constructing reality, of bringing sense to something that is obscure or unusual. The third influential book was Narrative Knowing arid the Human Sciences by Donald Polkinghorne (1988). While this book is wide-ranging in its scope, perhaps one of its most important features was the opening up of hc.rmen-eutic philosophy, in particular the work of Paul Ricoeur, to more widespread discussion within psychology, Ricoeur has developed an immense body of work on the cenlralily of narrative for meaning making. !n his classic work Time and Narrative, Ricoeur (1984) has argued that since we live in a temporal world we naecl to create narratives to bring order and meaning to the constantly changing flux. Further, not only do we create narratives about the world but also narrative is central to how we conceive of ourselves, to our identify. H is through narrative that we not only construct a particular connectedness in our actions but also distinguish ourselves from others. During the 1980s and 1990s, the study of narrative became much more extensive within various fields of psychology. Within personality and human development studies, Dan McAdams (1985) argued that narrative is central to our self-definition: We arc all tellers of tales. We each seek to provide our scattered and often confusing experiences with a sense of coherence by arranging the episodes of our lives into stories, (p. 11) Within clinical psychology, there was a movement towards the development of a form of narrative therapy (e.g., Mair, 1989; Neimeyer, 1995) that is based upon exploring alternative stories. Within health psychology, several researchers (e.g., Crossley, 1999; Murray, 1997a) argued that narrative is an everyday means of making sense of the disruption of illness. Of particular note, the study of narrative within psychology encouraged the growth of greater contact with the humanities (e.g., Fulford, 1999; Joy, 1997) and with the oilier social sciences (e.g., Maines, 1993). Dňftfiitioíi of Narrsíive According to narrative theory (e.g., Murray, 1999; Sarbin, 1986), we are born into a storied world, and we live our lives through the creation and exchange of narratives. A narrative can be defined as an organized interpretation of a sequence of events. This involves attributing agency to the characters in the narrative and inferring causal links between the events. In the classic Qualitative Psychology Narrative psychology 115 formulation, a narrative is an account with three components: a beginning, a middle and an end. Indeed, Bettina Becker (1999) has argued that in our world the number three has a special quality, For example, unlike the open-ended nature of a straight fine, a triangle is enclosed, finished. In the same way, a narrative offers an integrated account of an event. Unlike an open-ended piece of discourse, a narrative has a finished structure. The full dimensions of this structure may not be detailed in everyday conversation. Rather, depending upon the context, certain endings may be left unfinished, and it is the job of the audience/reader to complete the narrative. Since we live in a storied world, we can draw upon more established social narratives to explain an event or to complete a particular story. This is not a process of which we are always conscious. Ft-neüof! er řranTsíivfc Tlie primary function of narrative is that it brings order to disorder. In telling a story, the narrator is trying to organize the disorganized and to give it meaning. This is not a straightforward task. As Ricoeur (1987) says: The narrative ... is a synthesis of the heterogeneous. But concord cannot be without discord. Tragedy is paradigmatic for this: no tragedy is without complications, without fickle fate, without terrible and sad events, without irreparable error committed in ignorance or by mistake rather than through evil-mindedness. If then concord wins out over discord, surely it is the battle between them that makes the story, (p. 436) The ongoing tension continues as we try to give meaning to the various challenges to the order of our everyday life. Indeed, the tension intrinsic to narrative continues into the analysis of narrative accounts. This is often tentative and open to further challenge. The use of narrative is particularly pronounced in everyday understandings of disruption (e.g., Becker, 1997). We all encounter disruptions to our everyday routines. Such disruptions include personal problems, family problems, financial problems, and health problems. These challenges to our daily routines encourage attempts by us to restore some sense of order. Narrative is a primary means of restoring this sense of order. The classic experiment by Heider and Simmel (1944} is an illustration of what can be described as this human urge to narrative. In that experiment, participants were shown a sequence of pictograms of abstract shapes in different positions. When asked to describe the pictograms, the participants replied with short stories. Since Heider and Simmel were interested in how the participants attributed causal connections, they did not consider the structure of the stories they developed. Fortunately, some of these stones were included in their report of Ihe experiment, and it is apparent that although they were brief, the stories contained the basic elements of the classic narrative with a beginning, a middle and an end. Although we can use narratives to describe the movements of inanimate objects, such as in Heider and Simmei's experiment, it requires that we give those objects agency. Humans are action centres that strive within bounds to create their own worlds. They provide narrative accounts of their experiences that imply their role or lack of role in shaping these events. The converse of agency is suffering (Ricoeur, 1984), When we are denied the opportunity to express our agency, we experience suffering. Accounts of suffering reveal this restraint on our free agency. Suffering can be due to some personal misfortune, but it can also be due to social oppression that denies the opportunity for true agency. The need to restore a sense of order following disruption is especially pronounced in Western society, which is bounded by order and rationality. Oaylene Becker (1997) has argued that Western ideas about the life course emphasize linearity. Living in such a world, we try to make sense of inconsistencies. Further, when we try to explain our disruptions to another, we are particularly keen to emphasize our reasonableness. The central process of bringing order has been termed 'empäotmenť by Ricoeur (1984), to denote the organizing of a sequence of events into ä plot. This sequence of events can be brief or limitless. We can tell the story of going shopping or the story of the creation of the universe (cf. Polking-iiorne, 1996). The common theme is the attempt to give these events a narrative shape. Events do not just happen. In the narrative, there is an interconnected sequence that leads from start to finish. However, the event has ended before the narrator has started to construct a narrative. Freeman (1993) has alerted us to this process: Consider again the word 'recollection' itself: while the 're' makes >. reference to the past, 'collection' makes reference to a present act, ' an act . . . of gathering together what might have been dispersed or lost. (p. 40} In telling the story, the narrator is aware of the ending and constructs the account from there. In life, all narratives are provisional; they are subject to change as new information becomes available. It is not that the narrator is trying to mislead the listener but rather, from a more extended perspective, different pieces of information become available for the story. Narrative Identity Narrative not only brings order and meaning to our everyday life but, reflexively, it also provides structure to our very sense of selfhood. We tell 116 Qualitative Psychology stories about our lives to ourselves and to others. As such, we create a narrative identity. 'Subjects recognize themselves in the stories they tell about themselves' (Ricoeur, 1988: 247). We can hold a variety of narrative identities, each of which is connected to different social relationships. Each narrative identity not only connects us to a set of social relationships but also provides us with a sense of localized coherence and stability. At times of instability, we can make connections to other aspects of our narrative identities. It is through narrative that we begin to define ourselves, to clarify the continuity in our lives and to convey this to others. We are active agents who recall the actions we have achieved and also those that have been suppressed by others. Narrative enables us to describe these experiences and to define ourselves. In constructing a persona! narrative, we are selecting certain aspects of our lives and connecting them with others. This process enables us to assert that our Jives are not a disconnected sequence of events but have a certain order. This process of narrative identity formation is dynamic and occurs in a changing social and persona! context. The values attached to different experiences in that context influence the character of events recalled and thus the shape of the story to!d. As Ricoeur (1987; 437) emphasized, this indicates 'we learn to become the narrator of our own story without completely becoming the author of our life'. While we can telí our life story, the actual pattern our life takes and indeed the very structure of the story we teil are shaped by a multiplicity of social and psychological forces both conscious and unconscious (Hollway and Jefferson, 2000). Social Dimensions of Narraiives Narrative accounts are not emitted in a vacuum; rather, they are encouraged and shaped by a certain social context. Although the narrator tells the story, the character of the story told wiH depend upon whom the story is being told to, the relationship between the narrator and the audience, and the broader social and cultural context (Murray, 1997a). Thus, the study of narrative breaks down the traditional psychological/social distinction and develops a more complex psycho-social subject. The narrator is an active agent who is part of a social world. Through narrative, the agent engages with that world. Through narrative analysis, we can begin to understand both the narrators and their worlds. Although narrative is often considered in individua! or personal terms, we can also consider group, community or societal narratives. These are the narratives that particular collectives tell about themselves, their histories and their aspirations. In the same way as personal narratives are involved in the creation and re-creation of personal identities, these social narratives define the history of a collective and distinguish it from other collectives. Further, Narrative psychology 117 these collective narratives overlap with personal narratives such that individuals can define themselves as part of the group. In discussing narrative analysis, we should think about the level of analysis we are considering (Murray, 2000). Moreover, in analysing the personal narrative, we should Mtempt to consider the character of the broader social narrative within which it is being created. In sum, we are enmeshed in a world of narrative; we understand our world and ourselves through narrative. As such, the study of narrative provides the researcher with a means to understand how we make sense of the world and of ourselves. The meaning of different narratives is not always apparent and can be approached in different ways by different researchers. Narrative Research Collecting NatTSih'es The primary source of material for the narrative researcher is the interview. Unlike the traditional structured interview that has a detailed series of questions to be answered, the narrative interview is designed to provide an opportunity for the participant to give a detailed narrative account of a particular experience. The life-story interview is the most extended version of the personal narrative interview. Gerontológ is I s have particularly favoured this life-story approach as a means of exploring the experience of ageing (e.g., Barren et a)., 1996). As its name implies, the aim of the life-story interview is to encourage the participants to provide an extended account of their lives. The researcher will explain at the outset of the interview that the aim of the study is to learn about the person's life. While this may seem a simple invitation, the participant may, in practice, often be wary and uncommunicative at the outset. It is for this reason that the interviewer may need to meet with some participants on a number of occasions to win their confidence and to encourage them lo reflect on their life experiences. However, narratives are not just life stories in the most general sense but also stories about everyday experiences, especially disruptions of daily life. We can in the interview setting encourage participants to tell stories about particular experiences of change or disruptive episodes in their lives (cf. Hick, 2002). Given the time and the opportunity, participants are often very willing to provide extended narrative accounts of different experiences. .See Box 6.1 for examples of interview guides. The challenge is for researchers to convince the participants that they are interested in their narrative accounts. Thus, the researcher should reflect upon what the participants are saying and introduce supplementary questions designed to obtain clarification, such as 'Why do you think that 8 Qualitative Psychology Box 6.1 Sample interview guides 1. I would like you to tell me aboul yourself - where you were bom, where you grew up, thai sort of thing. You should nol be in any way inhibited about what lo say, but just teif me as much as possible about yourself. 2. I am interested in finding out what happened during the selection interview. You can begin at the time you left home for the meeting and just tell me as much as you can remember. is the case?' or 'Could you give an example or that?' Sometimes it may be useful to invite participants to a group meeting where they can share in tlii; íeläing of stones about an event. This focus group approach provides some participants with a greater sense of control and confidence (see Chapter 9 of this volume; also Wilkinson, 1998a; 1998b). These group interviews can lx-followed or supplemented with individual interviews. Another approach is for the researcher to provide the participant with a written list of the key issues to be discussed. This helps to alleviate any suspicions the participant might have that there are some trick questions to come. The interviewer can also use other methods such as encouraging the participants to keep a personal journal or to collect photographs or even to make a video. The aim is always to find a technique with which the participants are comfortable and which will allow them to develop their narrative account. Further, the researcher can analyse narrative materiál that is already available. For example, you can analyse published memoirs or films (e.g., Murray, 1997b). Since stories develop out of a particular social context and are told to a certain audience, it is important that such details are recorded when collecting narrative accounts. Mishler (1986) noted with reference to the importance of considering the interview setting: The interviewer's presence and form of involvement - how she or he listens, attends, encourages, interrupts, digresses, initiates topics, and terminates responses - is integral to a respondent's account. It is in this specific sense that a 'story' is a joint production, (p. 82) The researcher should collect background material about the centra! participants as well as details about the interviewer. Such information is important when we begin to analyse the narrative accounts. A useful strategy is for the researcher to keep a detailed log of each interview. This could include some basic demographic details of the Narrative psychology participant and when and where the interview occurred. Sometimes after the interview has ended and the tape recorder has been switched off, the participant will make some additional comments that can substantially influence the interpretation of the whole narrative. It is important that the researcher pay careful attention. After the interview has ended, the fvsearchers should record in their logs as much detail and commentary as they can recall about the interview. Even at this early stage the researcher should be considering what the key issues arising are and how the narrative is structured. Süme Logistical issues !t is important that care be taken in setting up the interview. The researcher should make some initial contact with the participants, explain the purpose of the study and obtain their consent. At this stage, they can discuss when it would be most convenient to return for a more extended interview and clarify where would be the most comfortable setting. Sometimes the participants are happy to be interviewed at home; other times they prefer to attend the researcher's office or another setting. H is important to remember that it is the participants' choice and to accommodate their preferences as much as possible. The researcher should practise using a tape recorder and test the quality of the recording. Sometimes the quality can be poor because of noise outside the interview room or the participants' speaking quietly. For these reasons, it is advisable to use an external microphone, if possible, the researcher should also attempt to ensure that the power is not interrupted during the interview by connecting the tape recorder to the electricity supply or by ensuring that extra batteries are available. It is a sign of respect to the participant for the researcher to ensure that the narrative is carefully recorded. Make sure to use the best available recording equipment and check that it works. Finding that the batteries fail halfway through the interview or discovering afterwards that the microphone did not pick up the participant's voice can be very frustrating. Often the novice interviewer will be apprehensive about using a tape recorder and think that this will inhibit the participant. Fortunately, the reverse is often the case. After the researcher has carefully explained the study and assured the participant of confidentiality, the participant is often very enthusiastic. Sometimes after an initial hesitation, the participants will proceed to talk at length about their various experiences. The very fact that they have an audience for the story can act as a spur to more sustained reflection. It is surprising that even when the interview is being videotaped and requires additional technical personnel, many people, once they have agreed to participate, will be only too generous with their time and will be surprisingly frank and revealing. It is for this reason that the researcher 120 Qualitative Psychology should treat the participant with the utmosl respect and courtesy. In addition, if the participant becomes distressed, the researcher should be prepared to stop the interview and, if necessary, suggest support opportunities. Afterwards, it is important that the interview be transcribed carefuiiy. It is a great advantage to have a professional transcriber, but this does not mean that the researcher does not have a role to play in the transcription process. Rather, the researcher should carefully review the transcript with the tape recording, correcting any errors in the transcription. This should be done as soon as possible after the interview, since it is easy to forget what the person had to say, especially when interviewing a number of people. There are different ways of preparing interview transcripts for analysis. This depends upon the analytic frame preferred by the researcher (cf. Chapter 7). In narrative analysis, the focus is on getting the main narrative account. The narrative transcription should include, where possible, exclamations, pauses and emphases. You can underline certain parts of the text that the participants have stressed in their speech ot add notes to mark such paralinguistics as sighs. The aim is to convey the detail and tenor of the story or stories. The transcription should aiso include the words of the researcher such that the character of the conversational exchange is apparent. We will see this more clearly in the example given below. Analysing Narratives The analysis of narrative accounts can be divided into two broad phases -the first descriptive and the second interpretive. A thorough reading of the transcribed narrative precedes both phases. In reading the narrative accounts, the aim is to familiarize oneself with both their structure and their content. A useful strategy is to prepare a short summary of the narratives that will identify the key features, such as the beginning, the middle and the end. The analysts can highlight key issues in the text and identify narrative linkages that connect different parts. They can also discern sub-plots within the broader narrative and consider connections between these. The sum-mary will highlight the particular features in which the researcher is interested. In reading across the summaries, it is then possible to begin to get an idea of what the main issues being raised are (Mishicr, 1986). It is through this process of close reading that a coding frame can be developed that can be applied to the various narratives. This coding frame is designed to capture the overall meaning of the narratives and the various particular issues raised within each. The second step is to connect the narrative with the broader theoretical literature that is being used to interpret the stoiy. Thus, the researcher goes beyond the descriptive phase to develop the interpretation. This requires a simultaneous familiarity with the narrative accounts and with the relevant literature such thai the one can begin to connect with the other. This phase Narrative psychology of the analysis can lead to labelling certain accounts as being of a certain type that illustrates their theoretical content. For example, we might be interested in how certain people handle particular crises in their lives. In the reading of the narratives, the central concern is how the narrators describe the various crises in their lives, how they draw on particular sources of support, and how they orient the story to the listener. Each story is examined for particular narrative elements - how the elements in the narrative arc linked together, what issues are emphasized and what metaphors are used. Role cf the Reader The process of narrative analysis is not a passive process. Rather, the researchers bring to the text certain assumptions and beliefs that they use to analyse the narrative. In discussing the process of reading a text, Ricoeur (1987) makes the same point: The meaning or the significance of a story wells up from the intersection of the world of the text and the world of the reader, (p. 430) Ricoeur (1991a) used the term 'appropriation' to describe the process of narrative interpretation. He defined this process as making one's own what has been alien. This is not a one-way process. Not only does the researcher firing to the narrative certain ideas but also, simultaneously, the narrator is trying to convince the audience of the character of his or her story. As Ricoeur (1991a) stresses: We play with a project, with an idea, we can equally be played. What is essential is the 'to and fro' {Hm und Her) of play. Piay is thereby close to dance, which is a movement that carries away the dancer, (p. 90) Thus, rather than imposing a framework and rather than simply describing the narrative account, narrative analysis requires that the analyst play with the account. In conducting the narrative analysis, it is important to be aware of what theoretical assumptions are guiding the analysis while at the same lime being open to new ideas and challenges. Hsrraiive Corden! sne; Structure A particular concern in narrative analysis is how the narrative is structured or organized. Various schemes have been developed to convey the temporal quality of narratives. The threefold classification scheme developed by (iergen and Gergen (1984) is a useful analytic tool, but it is important not to 122 Qualitative Psychology Narrative psychology apply ii in a schematic way but rather in ;i flexible marnici so as to encapsulate the various shifts in any narrative account. i;or example, the tragic narrative begins with a progressive structure, but then, despite struggle, tin1 central character is overcome and the narrative becomes regressive. This regression can be overcome by changing the broad interpretive dimensions that are being used to frame the event. For example, people who arc upwardly mobile in their career will probably present a progressive career narrative. However, if they are dismissed from their jobs they may develop a more regressive narrative unless they can redefine their goals and so continue to present a progressive narrative. This redefinition of goals, this turning point in a narrative, is similar to an epiphany. This is the moment in the account when the narrator sees the world in a different way. Conversely, a comedy is when a regressive narrative is transformed into a progressive narrative, as narrators redefine their values and realize the positive features of the changed life. in his analysis of the personal narratives of people with multiple sclerosis (MS), Robinson (1990) used this temporal scheme. He found that the MS narratives could be organized into the three broad categories. There were those who thought that their lives were ended due to the onset of MS {regressive narrative), those who thought that life had changed but was ongoing (stable narrative) and those who thought that the disease provided new opportunities (progressive narrative). Gee 0991) described the value of exploring the poetic structure of popular narrative accounts. He argued that verses are an intrinsic part of everyday narrative accounting and that poetry is merely a more developed form of that accounting. In particular, he was concerned about the use ol rhythm and metaphor in popular narratives. The study by Becker (1999) is an example of the successful use of this strategy to explore personal narratives. In reading through the pain narrative of an elderly person, she noted that it had a certain poetic quality. She was then able to recast the narrative account as a series of poetic stanzas that each had a similar structure. In recasting the narrative, the interviewer's questions are omitted and the text is organized into verses by the researcher. This form of analysis requires attention to the overall rhythm that underlies the narrative and the metaphors used to describe particular experiences. For example, the narrator may repeat certain phrases within her account (such as 'and then ľ) which provide a certain rhythm. The researcher can also consider the personal, interpersonal, group and societal contexts (Murray, 2000). The personal context is concerned with how narrative draws on the experience of the individual; the interpersonal and group context take into consideration the audience and the co-construction of narrative; and the societal context considers the broader social narratives which structure our everyday accounts. While it is difficult to integrate all these contextual levels into a single analysis, attention to one or the other may be particularly important in understanding the structure of certain narrative accounts. in this chapter, we will consider the structure of a personal narrative account and the value of different analytic strategies. We will begin by summarizing the case, proceed to how the narrative is structured, and then consider how the narrative is located within a particular social context. Although we will consider in detail only one case, it is useful in developing an argument to explore contrary cases. We will consider briefly a contrary case. This process enables the researcher to clarify particular strategies used by the participants in constructing their narratives. An Example: A Breast Cancer Story Tlie example is' taken from a study of how women handle the disruption of their lives as a consequence of having had breast cancer (Murray, 2002). We were interested in how the women integrated the disease into their everyday lives - how they gave it meaning. We were also interested in how these stories were constructed in a particular social and interpersonal context. In this sense, we were interested in how the broader social context intersects willi the personal narratives. All the women interviewed had had surgery for breast cancer. At their last check-up, there was no sign of recurrence and they had agreed to be Interviewed about the experience. The interviews took place in the women's homes or in the researcher's office. A young female research assistant who had no personal experience of breast cancer conducted them. For many of the women, the interview was an emotional experience. Several of them mentioned that they had had limited opportunity to discuss the operation with others. They feit that they had to present a strong face to their husbands and family members. The opportunity to talk freely about the event was largely welcomed. It is important that inexperienced researchers are briefed on the emotional intensity of some narrative interviews and that they have the opportunity to discuss the experience afterwards with their supervisor. We can begin by preparing a summary of each of the narrative accounts. There were certain commonalties in all of the stories that gave them the standard narrative structure: ]. Beginning: this was life before cancer. Different women emphasized particular aspects of their lives - family life, marriage, work, children, etc. The main thing was that cancer did not play a part in their lives. Some of the women tried to identify early experiences that might have contributed to the later development of the disease.