grounded theory, Kathy Charmaz, losses, staff, research, coding, qualitative analysis, challenging behaviour, grounded theory studies, theory studies, grounded theory study, interview data, constructivist approach, suffering, intimate partner violence, Explaining, data collection, qualitative methods, J. & Strauss, Further reading, Charmaz, practical guide, qualitative research techniques, qualitative research, quantitative research, Health Care for Women International, Anselm Strauss, Kathy Charmaz Boyd, Alison Tweed
10 Grounded Theory Methods for mental health
Practitioners Alison Tweed and Kathy Charmaz History of Grounded Theory Grounded theory as a qualitative methodology was developed by Barney G. Glaser and Anselm L. Strauss, two sociologists investigating the social processes of death and dying in hospital in the United States in the mid 1960s. Their book, The Discovery of Grounded Theory (Glaser & Strauss, 1967), explicated a rigorous set of inductively driven research strategies, designed to proceed systematically from a set of specific observations to a general conclusion or theory in order to describe and conceptualize people's views, actions and life experiences within the context in which they are lived. Their methods contrasted to dominant positivistic and quantitative forms of research at that time. Positivistic forms of research focused on experimentation and observation to empirically test and verify hypotheses as a means of describing the world in measurable variables. Glaser and Strauss highlighted qualitative methods of inquiry as a legitimate form of research in their own right. Grounded theory gained popularity in certain disciplines such as nursing over the next decade and was used in 1970s mental health research. Wilson's (1977) major mental health nursing study in the United States analysed over 200 hours of observation and interview data from a community setting for people diagnosed with schizophrenia. However, grounded theory did not begin to gain favour within psychology until another decade later. Psychologists initially conceptualized it for use within clinical and health psychology as a fruitful means of gaining alternative and in-depth perspectives of service users' experiences (e.g., Rennie et al., 1988). Glaser and Strauss each brought a unique set of assumptions to the development of grounded theory as a method: Glaser brought positivist notions of objectivity based upon his quantitative background whereas Strauss took a pragmatist stance, influenced by an interest in action, language and meaning. Whilst these epistemological differences Qualitative research methods
in Mental Health and Psychotherapy: A Guide for Students and Practitioners, First Edition. Edited by D. Harper and A.R. Thompson. C 2012 John Wiley & Sons, Ltd. Published 2012 by John Wiley & Sons, Ltd.
132 Alison Tweed and Kathy Charmaz could be argued as adding multiple dimensions to the method, they eventually led to Glaser and Strauss's alternative forms of grounded theory. Today, scholars view grounded theory's epistemological position as operating on a continuum from more positivist forms (Glaser, 1992) through post-positivist (Strauss & Corbin, 1990) to constructivist versions1 (Charmaz, 2006; for a review see Madill et al., 2000). Introduction to the Method The grounded theory method provides explicit strategies for Data collection
and analysis and aims to produce an inductively driven theory of social or psychological processes grounded in the material from which it was derived. Most typically, textual material has been the primary form of data, including interview transcripts, written documents and diaries. Glaser and Strauss (1967) outlined a number of characteristics of grounded theory, including the simultaneous involvement in collecting and analysing data, the development of analytic codes and categories and making comparisons between codes, concepts and categories. Starting with the data itself, developing a grounded theory can be conceptualized as a pyramid, where the raw data and the basic descriptive codes ascribed to the meaningunits of this data form the building block
s; the foundational base of the pyramid as it were. Focused codes and categories form the additional, less numerous blocks of the pyramid. These codes and categories conceptualize the basic codes beneath them and the pyramid builds towards its peak, each level denoting a higher, more sophisticated level of abstraction and interpretation. Finally comes the peak of the pyramid either representing a core category, encompassing all those codes and categories subsumed within it; or a theoretical conceptualization of the processes interpreted from the data. Here is the pinnacle of the analysis from which the storyline of the grounded theory can be conveyed to others. In addition to the principles outlined above, grounded theorists also use a series of analytical and reflexive strategies to aid the process of developing theory. The first of these is the constant Comparative Method
(Glaser & Strauss, 1967). Here, all elements of the analysis data, codes, categories and concepts are constantly compared within and between each other. This comparative process entails looking for similarities, differences and nuances between all the elements of the analysis in order to generate a more abstract understanding of the material. Using this comparative method is a dynamic non-linear process, requiring the researcher to stay open to new insights within the analysis. Assisting this process is the second strategy of memo-writing (Charmaz, 2006). Memo-writing is an intermediate stage between data collection and write-up and involves the detailed capturing of the researcher's thoughts, hunches, interpretations and decision-making throughout the analysis. We can view memo-writing as part of the 1 Whilst there is an accepted difference between constructivist and social constructionist perspectives within the United Kingdom, the term `constructivist grounded theory' has become widely used in the both the UK and US research literature
s. The worked example and approach presented in this chapter is consistent with a contemporary UK social constructionist approach.
Grounded Theory Methods 133
THEORY BUILDING FOCUSED CODING and CATEGORIZING MEMOWRITING CODING Data collection (interviewing) Sampling and recruitment of participants
Categories reaching SATURATION
Incomplete understanding raises questions
CONSTANT COMPARATIVE METHOD
Examples present within the data or from new data
THEORETICAL SAMPLINGResearch Question
Figure 10.1 Visual representation of a grounded theory. `audit trail' of grounded theory, enabling the recording of the analysis as it happens. We provide detailed examples of coding and memo-writing below to show how grounded theorists use these strategies. Finally, theoretical sampling is a strategy designed to sample new cases or data actively in order to develop, refine and elaborate the emerging grounded theory. Researchers engage in theoretical sampling after they have developed tentative analytic categories. Thus, theoretical sampling is unrelated to purposive sampling or representative sampling conducted at the beginning of the research process. Memos can alert the researcher to under-developed areas of theory, gaps or hypotheses requiring elaboration or testing. Here, the purely inductive emphasis of grounded theory alters because theoretical sampling contains both inductive and deductive elements. Typical theoretical sampling questions might be: `Whose voices are not represented by my tentative category?' or `Who do I need to speak to next to develop this theoretical category?' Ideally, theoretical sampling of new material continues until saturation is reached. Here, no new insights or development of the theory or properties of categories occur through the gathering of additional data. Depending on the scope of the research question, with interview transcripts saturation may occur after collecting considerable data or after checking key analytic categories through theoretical sampling of selected interviewees who possess the requisite knowledge to shed light on the properties of these categories (Charmaz, 2006).
134 Alison Tweed and Kathy Charmaz In order to aid the reader, Figure 10.1 presents a visual representation of the process of developing a grounded theory. The example uses interview transcripts as data and illustrates all the key elements from data gathering to dissemination. What Kinds of research questions
is Grounded Theory Most Suited To? Grounded theory's theoretical, epistemological and technical foundations position it well to investigate a broad range of open-ended research questions that focus on processes, patterns and meaning within context and that require the crucial examination of subjectivity of experience and thus lead researchers to begin inquiry from their research participants
' point of view. Grounded theory can be employed where existing theories or areas of research are under-defined or patchy and has the flexibility and sensitivity to be responsive to changing contexts and conditions (Henwood & Pidgeon, 2003). In other words, grounded theory methods foster specifying the implications of changes in social settings and situations where the studied phenomenon occurs as well as delineating the conditions under which it arises, is maintained, or varies. It is not surprising therefore that grounded theory has gained popularity as a method within mental health research and it has frequently been used to analyse the accounts of those individuals and groups typically perceived as `marginalized' (see, e.g., Boyd & Gumley, 2007; Charmaz, 2008; Charmaz & Rosenfeld, 2006; Karp, 1996). Further, a review of recently published grounded theory articles provides additional examples: clients' experiences of disengaged moments in psychotherapy (Frankel & Levitt, 2009), transition to motherhood for women with postnatal depression (Homewood et al., 2009), perceptions of the concept of recovery from schizophrenia (Noiseux & Ricard, 2008) and nurses' experiences of working in acute psychiatric settings (Chiovitti, 2008). However, grounded theory is not solely focused on individuals' interpretation of their experience but has a broader remit. Primarily, the goal of grounded theory is to develop an inductively driven theory of studying basic social or psychological processes (Glaser & Strauss, 1967) in which the researcher defines a fundamental process occurring in the setting or among the research participants and pursues researching it as the phenomenon of interest. The focus on social processes enables grounded theory to investigate how social structures
, situations and relationships influence patterns of behaviour, interactions and interpretations. This focus can include the impact of policies and services upon behaviour (e.g., Wuest et al., 2006). With apparent parallels in terms of researcher reflexivity and the self-reflection of therapists, psychotherapists have also seen grounded theory as a suitable method for psychotherapy research including systemic therapy (Burck, 2005) and psychoanalytic research (Anderson, 2006). What Kinds of Questions is Grounded Theory Not Suited To? Madill and Gough (2008) have provided a useful positioning framework, outlining where grounded theory as an analytic method can be placed in relation to the variety of qualitative approaches available. Within this framework, Madill and Gough
Grounded Theory Methods 135 highlight four areas in which qualitative methods can be organized in terms of their procedural categorization: discursive, thematic, structured and instrumental. Grounded theory has been described as thematic, alongside other methods such as Interpretative Phenomenological Analysis (Smith, 1996) and Thematic Analysis (Braun & Clarke, 2006). Following this argument, grounded theory would not be suitable for `testing out' existing theory or hypotheses, perhaps through the use of a priori domains or pre-designed coding schemes (structural methods). From a constructivist approach, neither is grounded theory a particularly suitable method for making truth statements or knowledge claims about an objective reality (Suddaby, 2006). Collecting Data: What Constitutes Data and How Much Should I Collect? Most commonly, grounded theory studies have been associated with the analysis of transcripts of interview data, typically gathered in a semi-structured format. However, this focus neglects the wide range of information that can be used as data and can provide novice researchers with a narrowed conceptualization of what constitutes data suitable for grounded theory analysis. Grounded theorists have used a number of sources of data including official records, letters and diary entries, fieldnotes based on observational work and focus group
material, in addition to interview transcripts (for a summary see Henwood & Pidgeon, 2003). How each researcher views and works with data depends on his or her epistemological position. From a post-positivist frame, the data will be seen factually as a representation of reality; from a constructivist frame, the data will have been constructed for a specific purpose and outcome and needs to be recognized as such (Charmaz, 2006). The amount of data to collect within a grounded theory study very much depends upon the research goal. Studies that are aiming to theorize will require greater amounts of data than those studies aiming to provide detailed descriptions of localized phenomena. Whilst pragmatic factors often play a part in determining the amount of data collected, it is the principle of saturation that should be the key consideration. Data collection ceases when theoretical categories are saturated (see previous section). In principle it sounds a relatively straightforward matter to determine whether categories are saturated, yet in actuality it is a complicated and challenging process and whilst it is easy to claim saturation, it is much more difficult to demonstrate it (Morse, 1995). Bowen (2008) argues that published grounded theory studies need to demonstrate the saturation process explicitly by providing clear descriptions and criteria used in the write-up. He provides an example of how this was achieved in his grounded theory investigation of community-based anti-poverty projects. How Might Participants and Service Users be Involved in Grounded Theory Studies? Undoubtedly, the most traditionally frequent form of participant involvement in a grounded theory study occurs during the data collection stage of the research. Here,
136 Alison Tweed and Kathy Charmaz the researcher actively seeks those individuals he or she believes can reveal something in relation to the phenomenon of interest. Yet to focus solely on the participant as the subjective focus of the research question as a means of eliciting data neglects those other areas where participants can successfully play a part. Charmaz (2006) discusses the notion of sensitizing concepts, a term originally described by Blumer (1969). Sensitizing concepts are seen as a starting point to grounded theory research through which the researcher generates initial ideas of interest, pays attention to guiding theoretical framework
s and becomes sensitized to asking particular types of questions, such as questions about identity or stigma. Participant involvement at this early stage can therefore be valuable in priming the researcher to key experiences of interest and providing a particular focus for the research. In an example of this, Boyd and Gumley (2007) undertook a grounded theory study with service users who had experiences of persecutory paranoia in order to develop an experiential perspective of this phenomenon. The authors were sensitized to the study by consultation with service users at the design stage who assisted in guiding the authors to particular areas of concern and relevance. However, only a small minority of grounded theory studies have demonstrated a broader range of service user involvement. One example is the study undertaken by Rose et al. (2004) investigating consumers' views of electroconvulsive therapy (ECT) through the use of a variety of data sources. Two of the authors had been recipients of ECT themselves and whilst the paper does not fully discuss the influence of these experiences upon the research process, they clearly sensitized the researchers and could well have improved both their understanding and engagement with the gathered data. Teram et al. (2005) also used grounded theory (alongside participatory action research) in a study investigating female survivors of child Sexual Abuse
and their experiences of physical therapy. They aimed to empower the research participants as a means of informing professional practice and training. How Does a Researcher Use Grounded Theory? What is coding? Coding is the first step of the analytic process and involves taking data apart. When coding, researchers break their data into analysable fragments and name these fragments. Thus, a code is a shorthand analytic label that a researcher defines. Grounded theorists aim to make their codes active, short, specific and spontaneous (Charmaz, 2006). Through coding, researchers gain an analytic handle on their material and an analytic point(s) of departure. Grounded theorists grapple with data and define their meanings through using codes. Coding serves the following objectives: 1. To engage researchers with their data without applying preconceived concepts; 2. To define what is happening in these data; 3. To compare, sort and categorize fragments of data; and 4. To begin connecting data with data, data with codes, and codes with codes.
Grounded Theory Methods 137 Constructivist grounded theorists acknowledge how researchers' and their research participants' social locations and perspectives and the situation of inquiry shape data (see also Clarke, 2005). Thus, researchers should account for their starting points and standpoints and analyse how they affect inquiry. Keeping codes active and specific helps the researcher accomplish this objective. Much research done in the name of grounded theory is descriptive and thematic. Thus, such research addresses topics as given rather than taking them apart and defining how they are constituted. Glaser and Strauss's (Glaser, 1978, 1998; Glaser & Strauss, 1967) early methodological strategies have always held potential for developing theoretical analyses. The grounded theory emphasis on analysing processes begins with using gerunds for coding. Because a gerund is the noun form of a verb, it preserves action by stating what people are doing, such as `defining', `explaining' or `accounting'. Gerunds also facilitate making connections between data and between codes. Thus, the researcher gains a handle on a greater portion of the data and links actions and events to reveal the story in the data, to invoke Strauss's focus.2 Coding with gerunds enables psychologists to: 1. Study emergent phenomena comparatively; 2. Make implicit meanings, actions and processes more visible and tangible; 3. Define relationships between inner mental processes
and external events; 4. Provide initial but modifiable conceptual handles for understanding and sorting large amounts of data; and 5. Construct successively more abstract theoretical categories and relationships. Inductive coding data has permeated qualitative research
, although coding with gerunds has not. Most researchers code for topics and themes at a considerably more general level than actions represented by gerunds. Topics or themes aid in sorting and synthesizing data but seldom break them apart as readily or make implicit relationships as visible as does coding for actions with gerunds (Charmaz, in press). Topics and themes tend to separate data into discrete units rather than reveal links between them. In addition to coding with gerunds, grounded theorists code small units of data. Line-by-line coding with gerunds is a heuristic device to bring the researcher into the data, interact with it and study each fragment of it (Box 10. 1). Line-by-line coding means giving each line in the data a short label to capture what the researcher defines is happening in each datum. Line-by-line coding is particularly helpful for analysing in-depth interviews or personal accounts. Using this type of coding during initial data collection assists researchers in defining directions to explore, helps them identify gaps in their data and spurs them to compare data and codes. Line-by-line coding with gerunds generates leads to pursue in subsequent data collection and ideas to check against previously collected data. 2 See the quotation from Hayes-Bautista in Charmaz (2000).
138 Alison Tweed and Kathy Charmaz
Box 10.1 Initial coding: line-by-line coding.
Acknowledging multiple reasons Repeating all-inclusive euphemism 'challenging' Delineating reasons stressing magnitude of losses: suffering rejection; identifying identity loss, family because of personality changes, children, home, job Losing one's life, way of being in the world Suffering massive loss; having alienated others Misbehaving cloaks loss Misreading behaviour (by staff) Acting up reflects loss Confronting loss constantly; surveying probable pasts Explaining all-encompassing loss Being aware of devastating losses; Losing one's prior life Having nothing left Comparing self with others; Explaining emotional response Understanding the magnifying of small slights; Forgetting to place patient's behaviour in context; staff bifurcating patient's biography
Excerpt 1, I: (3) The next question is thinking about why, um, patients with brain injury might exhibit challenging behaviour and you've, you've kind of talked a bit about that, you know, the processes in the brain. Is there anything else you want to add to that? P-D: There's so many reasons why a patient might display challenging behaviour. Loss of identity, you know, loss, they suffer huge losses, they might lose their family because of the change in personality, families just can't cope with, you know, this changing person. They might lose their children, they might lose their home, you know, they'll lose um, lose their job. I think, I don't think enough emphasis is placed on that fact that these people have lost their life. And basically through that they have perhaps experienced a huge range, huge range of losses and because of their behaviour they may have alienated people so that there is just this massive aspect of loss that they all have. And again, again I don't think that there are people that appreciate that, that they are doing it because they are acting up, or they are doing it because they want something when in effect, you know, they might just be constantly thinking about how much they've lost. I mean um, well you know . . . patient, well . . . patient has lost everything she lost her, you know she had a flat with her, well she was engaged she had a, you know, a really good job, fantastic social life. And she's got nothing now, and, you know, quite rightly so that, she's, you know, on days where she's feeling low and sees people going off doing things that she can't do. Her level of arousal is going to rise, so if you even say wait a minute to her, its important to her. And I think that's another thing that people forget it that, although it might seem a petty small request or something small that. People forget that to them [patients]
Grounded Theory Methods 139
Box 10.1 (Cont'd)
Acknowledging lack of life Pointing out staff's overlooking loss Putting self in patient's situation Understanding patient's frustration with varied causes of loss Connecting patient's losses with current mental health Understanding loss of autonomy, dignity Placing patient's anger in context Failing to grasp meanings of loss Lacking empathystaff
that's, that's important to them and they haven't got much of a life, um (.) I do think that once people have been here a long time, I think some people do forget that these patients have lost everything. And quite frankly I think I'd probably want to be as awkward as I could if it meant that I would get somebody to talk to me. Um, if they can't communicate, you know quite often they'll lose their ability to speak, you know, they're going to get frustrated with that. If they've lost their ability to walk, you know its just so many, so many different aspects that could, you know, it can impact on their mental health if they had issues before, or they might, you know, they might become depressed (2). Its just so, so, physical issues, things that they can't do. You know, its somebody who was quite proud, you know, quite an independent person who can't wash themselves, is going to get pissed off with it. Something that people don't, aren't, don't get a grasp of through just working with patients every day. I don't think people really, I don't think the empathy is always there. . . .
The coding in Box 10.1 is a secondary analysis of data from a study of staff in hospital for people with brain impairment (Stewart, 2007). The interviewer aimed to discover which kinds of patient behaviour staff found to be troubling. In keeping with grounded theory logic, she adopted the staff 's term, `challenging behaviour', and asked questions to break open this institutional euphemism. She crafted well-designed questions to find out what challenging behaviour meant to the interview participants. Her openended interviewing style allowed interviewees to concentrate on what they saw as most significant while she simultaneously explored the properties of challenging behaviour and the conditions under which staff defined it. Note that the codes in Box 10.1 use gerunds and demonstrate actions. The interview excerpt in Box 10.1 draws on three sources of actions: those of Participant D (P-D), her descriptions of actions of brain-injured patients in the hospital where she works, and her views of the actions of other people (staff) toward patients' actions. What is Participant D doing? The codes define her actions as `acknowledging multiple reasons', `repeating the all-inclusive euphemism, "challenging behaviour"', `delineating reasons' for this behaviour, `explaining all-encompassing loss' and `putting self in patient's situation'. Participant D explains these patients' experience in codes such as `suffering rejection',
140 Alison Tweed and Kathy Charmaz `suffering massive loss', `having nothing left' and `being aware of devastating losses'. When she considers patients' challenging behaviour in the hospital, Participant D's statements and therefore the codes reveal another line of action such as `misbehaving cloaks loss', `misreading behaviour' (by staff), `forgetting to place patient's behaviour in context', `staff bifurcating patient's biography', `failing to grasp meanings of loss' and `lacking empathy' (for guidelines for coding see Charmaz, 2006; Corbin & Strauss, 2008). All these codes suggest areas that the researcher could develop and check in subsequent interviews or observations. Might another analyst come up with different codes? Yes, our perspectives and social locations affect how we code. Different researchers coding the same data may generate new insights. After studying their initial codes, researchers can then treat their most compelling and frequent codes as focused codes to sort, synthesize and analyse large batches of data (Charmaz, 2006). What does memo-writing involve? Memo-writing is the pivotal analytic stage between coding and writing the first draft of a paper or chapter. Memo-writing occurs from the beginning of data collection and proceeds throughout the research process. Early memos are partial, tentative and exploratory, filled with empirical information to check and analytic questions to pursue (see the questions in the early sample memo below, `Explaining All-Encompassing Loss'). Later memos are more precise, abstract, sophisticated and conceptually robust, and may demonstrate relationships between theoretical categories as well as analyse a single category such as becoming marginalized. By subjecting key codes to further analysis, grounded theorists probe their data and examine how these codes hold up as tentative categories. Further checks occur as grounded theorists seek increasingly specific data to test their emerging analyses of the category or categories they are developing. Box 10.2 Sample Memo Explaining All-Encompassing Loss Explaining all encompassing loss means making explicit unknown or forgotten meanings of the magnitude of patients' loss. Explaining here means pointing out types of loss patients have experienced, delineating their extent, and making these losses known and understood. Explaining all-encompassing loss means taking the patient's perspective and looking at what is lost. Loss resides in the chasm between the life once lived and current institutional existence. Participant D points out, (Continued )
Grounded Theory Methods 141 Box 10.2 (Cont'd) "I think, I don't think enough emphasis is placed on that fact that these people have lost their life. And basically through that they have perhaps experienced a huge range, huge range of losses and because of their behaviour they may have alienated people so that there is just this massive aspect of loss that they all have." Thus, loss can result in spiraling consequences. Awareness of losing one's life, one's way of being in the world causes patients enormous suffering that they may express through frustration, anger, and aggression, which leads to being rejected and further suffering, and subsequently more acting up. Misbehaving cloaks loss, and then, staff misread the patient's behavior. Conditions that exacerbate this process include the nature of the patient's impairment and its relative visibility, the extent to which it complicates daily life, and the institutional situation itself. To what extent does being subject to this situation impart messages that lapses in self-control are routine events and therefore desensitize patients as well as staff to troublesome behavior? Explaining all-encompassing loss not only asks the listener to envision losses, but also to envision who a patient was before experiencing brain impairment. Thus, staff would gain a different image of the patient than that of the person they encounter in their daily work. Explaining links the past with the present and accounts for the present. (In contrast to Participant D's accounting for the present by looking at the past, Participant B considers impairment but concentrates on the present, not the magnitude of loss, nor the suffering it may cause.) Participant D adopts the role of the teacher who elucidates for the interviewer why patients act as they do. To what extent can or does she make her views known and heard? How does she deal with co-workers who fail to grasp these meanings of loss? By explaining all-encompassing loss, Participant D presents herself as empathetic, insightful, and different separate? from co-workers. How does her empathy alter relationships with patients? How and when do her insights affect her work with patients? What, if any, are the implications of setting oneself apart from co-workers in this setting? At every stage of memo-writing, grounded theorists analyse their data and thus move beyond description and summaries of data. Memo-writing advances comparative processes of comparing data with data, data with code, code with code, code with category, and category with category. Memo-writing also prompts a researcher to define the code, delineate its properties and to specify conditions when this code is or is not manifest. Note that the memo above specifies conditions and outlines how this code fits into an overall process that integrates other codes. Does the memo capture empirical reality?
142 Alison Tweed and Kathy Charmaz Through memo-writing, researchers may also explicate research participants' implicit assumptions and show how the codes reveal them. Writing memos spurs making discoveries because researchers develop fleeting ideas and define phenomena that they had not noticed before coding. Several other memo-writing suggestions help: 1. Title memos: make sorting, synthesizing, and storing easy; 2. Keep the analysis grounded: construct definitions and properties of codes or cate- gories from data; 3. Show processes: delineate conditions under which a code or category emerges, is maintained or changes; 4. Include evidence: insert data excerpts that generate or support a code or category; and 5. Track ideas: write memos whenever an idea arises. Memos give grounded theorists an analytic understanding of processes, substance for papers and chapters, tentative categories to interrogate further through theoretical sampling, and material for future projects. What Makes for a Better Quality Grounded Theory Study? Similar to all forms of research, grounded theory studies vary in quality and unfortunately many published examples offer little more than a glorified description of an experience or phenomenon. In psychology, key tenets in determining the quality of qualitative research studies relate to the adaptation of the scientific canons of validity, reliability and generalizability. Within grounded theory, Strauss and Corbin (1990, 1998) also guard against the unquestioning adherence to the three scientific canons as applied to quantitative research. They argue that these canons `require redefinition in order to fit the realities of qualitative research and the complexities of social phenomena' (Strauss & Corbin, 1990, p. 250). To evaluate grounded theory studies, Strauss and Corbin pose questions about the data, research process and empirical grounding of the findings. Their questions relate to systematic and transparent handling of the data, the level and development of categories and processes, and the significance of any theoretical findings. Charmaz (2006) also highlights evaluative criteria for grounded theory studies, classifying the areas of credibility, originality, resonance and usefulness. Scholars can use these criteria to evaluate studies by Charmaz (1991), Qin and Lykes (2006) and Shrock and Padavic (2007). Yet, even within these useful frameworks, researchers' epistemological positions shape their different emphases on various criteria. Madill et al. (2000) have used an example of a grounded theory study to compare how different criteria may be applied when alternative epistemological positions are taken: post-positivism and constructivist. They conclude that research needs to be `evaluated by the standards entailed by its own logic of justification' (Madill et al., 2000, p. 17).
Grounded Theory Methods 143 Future Directions for Grounded Theory The increasing popularity of grounded theory since its conception in the late 1960s has naturally led to an evolution in terms of grounded theory's epistemological underpinnings, emphasis and methodology. Constructivist versions of grounded theory have been developed (see Charmaz, 2006), altering the emphasis of the approach from discovery of theory to generation, accepting the interplay and connectivity between the researched, researched and interpretations made. Moves towards pragmatism, pleuralism and mixed methods (Madill & Gough, 2008) have enabled grounded theorists to use other methodologies as a complement in order to address broad-ranging research questions. In addition, new frameworks for analysis and theory-building such as situational analysis (Clarke, 2003, 2005) and fractal concept analysis (Wasserman et al., 2009) have assisted researchers in moving away from descriptive accounts. Without doubt, this new generation of grounded theorists will continue to develop grounded theory in interesting and innovative ways. As with epistemology and emphasis, the forms of media suitable for grounded theory analysis have also evolved in recent times, providing researchers with alternatives to the commonly used interview method. Grounded theorists have used focus groups to generate material (Henwood & Pidgeon, 2003), reflective commentary of events using Interpersonal Process Recall (Rennie, 2005) and increasingly recognize that visual media may be a suitable form of data for qualitative researchers to analyse (Gleeson et al., 2005). Within mental health research, these new directions alongside greater service user involvement throughout the entire research process provide not only challenges but opportunities to grounded theory researchers. It is an exciting time to engage in grounded theory research, not only to elucidate and situate the social and organizational processes service users experience, but to strengthen the connections between the academic and clinical worlds. Acknowledgement The authors would particularly like to thank Dr Inga Stewart, Senior Clinical Psychologist, for allowing us to use some of her interview data (with the participants' permission) for the worked examples in the coding and memo-writing sections of the chapter. It is very much appreciated. References Anderson, J. (2006). Well-suited partners: Psychoanalytic research and grounded theory. Journal of Child Psychotherapy, 32, 329348. Blumer, H. (1969). Symbolic interactionism. Englewood Cliffs, NJ
: Prentice-Hall. Bowen, G.A. (2008). Naturalistic inquiry and the saturation concept: A research note. Qualitative Research, 8, 137152.
144 Alison Tweed and Kathy Charmaz Boyd, T. & Gumley, A. (2007). An experiential perspective on persecutory paranoia: A grounded theory construction. Psychology and Psychotherapy: Theory, Research and Practice, 80, 122. Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77102. Burck, C. (2005). Comparing qualitative research methodologies for systemic research: The use of grounded theory, discourse analysis
and narrative analysis. Journal of Family Therapy, 27, 237262. Charmaz, K. (1991). Good days, bad days: The self in chronic illness and time. New Brunswick
, NJ: Rutgers University Press. Charmaz, K. (2000). Teachings of Anselm Strauss: Remembrances and reflections. Sociological Perspectives, 43, S163S174. Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis. London: Sage. Charmaz, K. (2008). Views from the margins: Voices, silences, and suffering. Qualitative Research in Psychology, 5, 718. Charmaz, K. (2009). Shifting the grounds: Constructivist grounded theory methods for the twenty-first century. In J. Morse, P. Stern, J. Corbin, B. Bowers, K. Charmaz & A. Clarke. Developing grounded theory: The second generation (pp. 127154). Walnut Creek, CA: Left Coast Press. Charmaz, K. (in press). Grounded theory methods in social justice research. In N.K. Denzin & Y.E. Lincoln (Eds.)Handbook of qualitative research (4th edn). Thousand Oaks, CA: Sage. Charmaz, K. & Rosenfeld, D. (2006). Reflections of the body, images of self: Visibility and invisibility in chronic illness and disability. In D.D. Waskul & P. Vannini (Eds.)Body/embodiment: Symbolic interaction and the sociology of the body (pp. 3550). London: Ashgate. Chiovitti, R.F. (2008). Nurses' meaning of caring with patients in acute psychiatric hospital settings: A grounded theory study. International Journal
of Nursing Studies, 45, 203223. Clarke, A.E. (2003) Situational analysis: Grounded theory mapping after the postmodern turn. Symbolic Interaction, 26, 553576. Clarke, A.E. (2005). Situational analysis: Grounded theory after the postmodern turn. Thousand Oaks, CA: Sage. Corbin, J.M. & Strauss, A.C. (2008). Basics of qualitative research: Techniques and procedures for developing grounded theory (3rd edn). Los Angeles
, CA: Sage. Frankel, Z. & Levitt, H.M. (2009). Clients' experiences of disengaged moments in psychotherapy: A grounded theory analysis. Journal of Contemporary Psychotherapy, 39, 171186. Glaser, B.G. (1978). Theoretical sensitivity. Mill Valley, CA: Sociology Press. Glaser, B.G. (1992). Basics of grounded theory analysis. Mill Valley, CA: Sociology Press. Glaser, B.G. (1998). Doing grounded theory: Issues and discussions. Mill Valley, CA: Sociology Press. Glaser, B.G. & Strauss, A.L. (1967). The discovery of grounded theory. Chicago: Aldine. Gleeson, K., Archer, L., Riley, S. & Frith, H. (2005). Visual methodologies. Qualitative Research in Psychology, 2, 187198. Henwood, K.L. & Pidgeon, N.F. (2003). Grounded theory in psychological research
. In P. Camic, J. Rhodes & L. Yardley (Eds.) Qualitative research in psychology: Expanding perspectives in methodology and design (pp. 131155). Washington, DC
: APA Publications.
Grounded Theory Methods 145 Homewood, E., Tweed, A., Cree, M. & Crossley, J. (2009). Becoming occluded: The transition to motherhood of women with post-natal depression. Qualitative Research in Psychology, 6, 117. Karp, D. (1996). Speaking of sadness: Depression, disconnection, and the meanings of illness. New York: Oxford University Press. Madill, A. & Gough, B. (2008). Qualitative research and its place in psychological science. Psychological Methods, 13, 254271. Madill, A., Jordan, A. & Shirley, C. (2000). Objectivity and reliability in qualitative analysis: Realist, contextualist and radical constructivist epistemologies. British Journal of Psychology
, 91, 120. Morse, J.M. (1995). The significance of saturation. Qualitative Health Research, 5, 147149. Noiseux, S. & Ricard, N. (2008). Recovery as perceived by people with schizophrenia, family members
and health professionals: A grounded theory. International Journal of Nursing Studies, 45, 11481162. Qin, D. & Lykes, M.B. (2006). Reweaving a fragmented self: A grounded theory of selfunderstanding among Chinese women students in the United States of America. International Journal of qualitative studies
in Education, 19, 177200. Rennie, D.L. (2005). The grounded theory method: Application of a variant of its procedure of constant comparative analysis
to psychotherapy research. In C.T. Fischer (Ed.) Qualitative research methods for psychologists: Introduction through Empirical Studies
(pp. 5978). London: Academic Press. Rennie, D., Phillips, J.R. & Quartaro, G.K. (1988). Grounded theory: A promising approach to conceptualisation in psychology. Canadian Psychology, 29, 139150. Rose, D., Fleischmann, P. & Wykes, T. (2004). Consumers' views of electroconvulsive therapy: A qualitative analysis. Journal of Mental Health, 13, 285293. Schrock, D.P. & Padavic, I. (2007). Negotiating hegemonic masculinity in a batterer intervention program. Gender and Society, 21, 625649. Smith, J.A. (1996). Beyond the divide between cognition and discourse: Using interpretative phenomenological analysis in health psychology. Psychology and Health, 11, 261271. Stewart, I. (2007). Attributions regarding challenging behaviour within an acquired brain injury setting: A grounded theory analysis. Unpublished thesis: Leicester University. Strauss, A. & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage. Strauss, A. & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd edn). Newbury Park, CA: Sage. Suddaby, R. (2006). From the editors: What grounded theory is not. Academy of Management Journal, 49, 633642. Teram, E., Schachter, C.L. & Stalker, C.A. (2005). The case for integrating grounded theory and participatory action research: Empowering clients to inform professional practice. Qualitative Health Research, 15, 11291140. Wasserman, J.A., Clair, J.M. & Wilson, K.L. (2009). Problematics of grounded theory: Innovations for developing an increasingly rigorous qualitative method. Qualitative Research, 9, 355381. Wilson, H.S. (1977). Limiting intrusion: Social control of outsiders in a healing community. Nursing Research, 26, 105111. Wuest, J., Ford-Gilboe, M., Merritt-Gray, M. & Lemire, S. (2006). Using grounded theory to generate a theoretical understanding of the effects of child custody policy on women's health
146 Alison Tweed and Kathy Charmaz promotion in the context of intimate partner violence. health care
for Women International, 27, 490512. Further reading and useful website Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis. London: Sage. Corbin, J. & Strauss, A. (2008) Basics of qualitative research techniques
and procedures for developing grounded theory (3rd edn). Los Angeles: Sage. Grounded Theory Institute: www.groundedtheory.com.
A Tweed, K Charmaz