Grounded Theory Methods in Social Justice Research
Director, Faculty Writing Program
Professor of Sociology
Sonoma State University
1801 E. Cotati Avenue
Rohnert Park, CA 94928
Qualitative research has long attracted researchers who hope that their studies will matter in the public arena as well as in their disciplines. Yet many qualitative studies have been conducted that posed intriguing intellectual questions, addressed an interesting population, or explored an understudied phenomenon without raising explicit questions concerning social justice or policies that address inequities. Such studies could often be taken a step or two further to explicate and explore social justice issues and subsequently reframe discussion of the studied phenomenon. What does social justice research entail? How can qualitative researchers move in this direction? What tools do they need?
When I speak of social justice inquiry, I mean studies that attend to inequities and equality, barriers and access, poverty and privilege, individual rights and the collective good, and their implications for suffering. Social justice inquiry also includes taking a critical stance toward social structures and processes that shape individual and collective life. I cast a wide net here across areas and levels of analysis in which questions about social justice arise. I include micro, meso, and macro levels of analysis, the local and global, as well as relationships between these levels. In the past, much social justice inquiry has assumed studying macro structural relationships but issues concerning social justice occur in micro situations and meso contexts, as well as in macro worlds and processes. Social scientists can study how the macro affects the micro and how micro processes also influence larger social entities. Global, national, and local social and economic conditions shape and are shaped by collective and individual meanings and actions. Yet when, how, and to what extent these conditions affect specific groups and individuals may not be fully recognized.
This chapter builds on my argument in the third edition of the Handbook of Qualitative Research (Denzin and Lincoln, 2005): qualitative researchers can use grounded theory methods to advance social justice inquiry. These methods begin with inductive logic, use emergent strategies, rely on comparative inquiry, and are explicitly analytic. All these attributes give social justice researchers tools to sharpen and specify their analyses that will increase the analytic power and influence of their work while simultaneously expediting the research process. Grounded theory methods not only offer social justice researchers tools for developing innovative analyses but also for examining established concepts afresh. To mine the largely untapped potential of grounded theory methods for social justice inquiry, social justice researchers need to understand the logic of the method, the development of its different versions, their epistemological roots, and how they might use it.
Research in the area of social justice addresses differential power, prestige, resources, and suffering among peoples and individuals. It focuses on and furthers equitable resources, fairness, and eradication of oppression (Feagin, 1999). Some reports in social justice inquiry begin with an explicit value stance and an agenda for change (see, for example, these grounded theory studies: Karabanow, 2008; Nack, 2008; Sakamoto, Chin, Chapra, and Ricciar, 2009; Ullman and Townsend, 2008). Other research reports often convey a taken-for-granted concern with social justice (e,g, Dumit, 2006; Foote-Ardah, 2003; Frohmann, 1991, 1998; Gagné, 1996; Hyde and Kammerer, 2009; Jiménez, 2008; Lutgen-Sandvik, 2008; Lio, Melzer, and Reese, 2008; Mevorach, 2008; Moore, 2005; Swahnberg, Thapar-Björkert, and Berterö 2007; Tuason, 2008; Veale and Stavrou, 2007). Still other authors indicate that they chose a controversial topic that has social justice implications because it could illuminate a theoretical problem (Einwohner and Spencer, 2005; Ogle, Eckman, and Leslie 2003; Spencer and Triche, 1994). Researchers may, however, begin their studies with an interest in a social issue rather than an impassioned commitment to changing it (Wasserman and Clair, 2010). Yet, the very process of witnessing their participants’ lives and analyzing their data may elicit concerns about social justice that they had not understood earlier or anticipated.
Many researchers hold ideals of creating a good society and a better world and thus pursue empirical studies to further their ideas. For those who identify themselves as social justice researchers, “shoulds” and “oughts” are part of the research process and product. Claiming an explicit value position and studying controversial topics can result in having one’s work contested. Hence, some researchers remain silent about their value commitments and instead choose to frame their studies in conceptual terms, rather than social justice concerns.
Grounded theory is a method of qualitative inquiry in which data collection and analysis reciprocally inform and shape each other through an emergent iterative process. The term, “grounded theory,” refers to this method and its product, a theory developed from successive conceptual analysis of data. Researchers may adopt grounded theory strategies while using a variety of data collection methods. Grounded theory studies have frequently been interview studies and some studies have used documents (Clarke, 1998; Einwohner and Spencer, 2005; Mulcahy, 1995; Star, 1989) or ethnographic data (e.g. Caspar, 1998; Thornton, 2006; Wasserman and Clair, In press; Wolkomir, 2001, 2006). It is often difficult, however, to discern the extent to which researchers have engaged grounded theory strategies (Charmaz, 2007, In press; Timmermans and Tavory, 2007).
The strategies of grounded theory provide a useful toolkit for social justice researchers to employ. Grounded theory practice consists of emergent research decisions and actions that particularly fit social justice studies. The grounded theory emphases on empirical scrutiny and analytic precision fosters creating nuanced analyses of how social and economic conditions work in specific situations, whether or not researchers take their work into explicit theory construction (see, for example, Ball, Perkins, Hollingsworth, Whittington, and King, 2009; Dixon, 2007; Jackson- Jacobs, 2004; Lazzari, Ford, and Haughey, 1996; Sixsmith,1999; Speed and Luker, 2006). Such analyses not only contribute to knowledge, but also can inform those practices and policies that social justice researchers seek to change.
Researchers can learn how to use grounded theory guidelines and put them to use for diverse research objectives, including interrogating social justice questions. To date, few grounded theory studies in social justice inquiry demonstrate theory construction. Many, however, show how grounded theory guidelines have sharpened thematic analyses. This chapter aims to clarify the method and its evolution, illuminate how researchers have used specific grounded theory guidelines, and demonstrate how this method complements social justice inquiry.
The constructivist revision of Glaser and Strauss’s (1967) classic statement of grounded theory assumes that people construct both the studied phenomenon and the research process through their actions. This approach recognizes the constraints that historical, social, and situational conditions exert on these actions and acknowledges the researcher’s active role in shaping the data and analysis. The constructivist version is particularly useful in social justice inquiry because it (1) rejects claims of objectivity, (2) locates researchers’ generalizations, (3) considers researchers and participants’ relative positions and standpoints, (4) emphasizes reflexivity, (5) adopts sensitizing concepts such as power, privilege, equity, and oppression, and (6) remains alert to variation and difference (see Bryant and Charmaz, 2007; Charmaz, 2006, 2009; Clarke, 2005, 2007). Nonetheless, adopting strategies common to all versions of grounded theory will advance social justice studies and, therefore, I discuss works that use all versions of grounded theory.
My discussion relies on a selective review of grounded theory studies in social justice inquiry. The burgeoning number and range of relevant works across disciplines and professions precludes a comprehensive review. The selected studies (1) show connections between social justice inquiry and grounded theory, (2) reveal debates concerning grounded theory, and (3) demonstrate ways to use this method. Like other qualitative research projects, many social justice studies indicate having used grounded theory strategies only for coding and confuse developing thematic topics for theoretical categories. One purpose of this chapter is to help researchers to make informed choices about when, how, and to what extent they adopt grounded theory logic and strategies. Grounded theory strategies can help scholars with diverse pursuits without necessarily developing a grounded theory. The point is to make clear decisions and to be aware of their implications.
The Logic of Grounded Theory
Grounded theory is a method of social scientific theory construction. As Glaser and Strauss (1967) first stated, the grounded theory method consists of flexible analytic guidelines that enable researchers to focus their data collection and to build middle-range theories. These guidelines emphasize studying processes in the field setting(s), engaging in simultaneous data collection and analysis, adopting comparative methods, and checking and elaborating our tentative categories. We grounded theorists begin with a systematic inductive approach to inquiry but do not stop with induction as we subject our findings and tentative categories to rigorous tests.
Fundamentally, grounded theory is an iterative, comparative, interactive, and abductive method (Bryant and Charmaz, 2007; Charmaz, 2006, 2007, 2008e; Charmaz and Henwood, 2008). The grounded theory method leads researchers to go back and forth between analysis and data collection because each informs and advances the other. By asking analytic questions during each step in the iterative process, the researcher raises the abstract level of the analysis and intensifies its power. Using comparative methods throughout the analytic—and writing—processes sharpens a researcher’s emerging analysis. Moreover, using a comparative approach in an iterative process keeps grounded theorists interacting with their data by asking analytic questions of these data and emerging analyses. Thus the strength of grounded theory not only resides in its comparative methodology but moreover, in its interactive essence (Charmaz, 2006, 2007, 2008a, 2008e, 2009). The method encourages researchers to become active, engaged analysts. Abductive reasoning keeps researchers involved. As grounded theorists, we engage in abductive reasoning when we come across a surprising finding during inductive data collection. Then we consider all possible theoretical accounts for this finding, form hypotheses or questions about them, and subsequently test these explanations with new data (Peirce, 1958; Reichert, 2007; Rosenthal, 2004). Abductive reasoning advances theory construction.
What does using the method involve? Grounded theory prompts us to study and interact with our data by moving through comparative levels of analysis. First, we compare data with data as we develop codes; next we compare data with codes; after that, we compare codes and raise significant codes to tentative categories; then we compare data and codes with these categories; subsequently we treat our major category(ies) as a concept(s), and last, we compare concept with concept, which may include comparing our concept with disciplinary concepts. The analytic comparisons we make during our current phase of inquiry shape what we will do in the next phase and cannot be ascertained beforehand. The method prompts us to interact with our participants, data, codes, and tentative categories. Through these interactions, our nascent analyses emerge and take form (Charmaz, 2006; 2007; 2008b; 2008c, 2008e).
This comparative, interactive process of inquiry leads us to move back and forth between data collection and analysis as each informs the other (Charmaz and Henwood, 2007). The grounded theory emphasis on theory construction influences how we interact with our participants and the questions we bring to the empirical world (see Charmaz, 2009a). The comparisons sharpen our analyses and the iterative data collection allows us to test our ideas and to check our emerging theoretical concepts. Grounded theorizing involves imaginative interpretations and rigorous examination of our data and nascent analyses (Charmaz, 2006; Kearney, 2007; Locke, 2007) Our systematic scrutiny not only increases analytic precision but also keeps us close to the data and, thus, strengthens our claims about it. Such an approach helps social justice researchers make their work visible and their voices heard.
In short, the logic of grounded theory involves fragmenting empirical data through coding and working with resultant codes to construct abstract categories that fit these data and offer a conceptual analysis of them (Charmaz, 2006; Glaser, 1978, 1998). Grounded theorists start with empirical specifics to move toward general statements about their emergent categories and the relationships between them. This approach allows social justice researchers to address problems in specific empirical worlds and to theorize how their categories may apply to other situations and iniquities (Dixon, 2007; Lutgen-Sandvik, 2008; Rivera, 2008; Shelley, 2001; Wolkomir, 2003).
Grounded Theory Strategies in Social Justice Inquiry
The analytic power of grounded theory offers qualitative researchers distinct advantages in pursuing social justice inquiry. Five grounded theory strengths make it a particularly useful toolkit for social justice researchers. First, this method contains tools for analyzing and situating processes. Thus, the logic of grounded theory leads to: (1) defining relevant processes, (2) demonstrating their contexts, (3) specifying the conditions in which these processes occur, (4) conceptualizing their phases, (5) explicating what contributes to their stability and/or change, and (6) outlining their consequences. Adopting this logic can help social justice researchers attend to the construction of inequities and how people act toward them. Thus grounded theory logic can lead researchers to make explicit interpretations of what is happening in the empirical world and to offer an analysis that depicts how and why it happens.
Second, grounded theory can aid researchers in explicating their participants’ implicit meanings and actions (see, for example, McPhail and DiNitto, 2005). A task for social justice researchers is to see beyond the obvious. The most significant meanings and actions in a field setting are often implicit. Successive, meticulous grounded theory analysis can help researchers to define implicit meanings and actions and to theorize tentative but plausible accounts of them. Subsequently, the grounded theory guideline of checking hunches and conjectures encourages researchers to subject their tentative ideas to rigorous scrutiny and to develop more robust analyses.
Third, the purpose of grounded theory is to construct middle range theory from data. Hence, grounded theory can aid social justice researchers to increase the abstract level of conceptualization of their analyses. Social justice researchers can then identify the conditions under which their categories emerge, specify relationships between these categories, and define the consequences. Thus, they can build complexity into their analyses that challenge conventional explanations of the studied phenomenon.
Fourth, the constructivist version of grounded theory attends to context, positions, discourses, and meanings and actions and thus can be used to advance understandings of how power, oppression, and inequities differentially affect individuals, groups and categories of people. Last, but extremely significant, grounded theory methods provide tools to reveal links between concrete experiences of suffering and social structure, culture, and social practices or policies (Charmaz, 2007; Choi and Holroyd, 2007; Einwohner and Spencer, 2005; Rier, 2007; Sandstrom, 1990; 1998).
To date, few researchers who adopt grounded theory methods have explicitly framed their work as contributions to social justice inquiry or made social justice issues a central focus (but see, Mitchell and McCuster, 2008; Sakamoto et al., 2009; Tuason, 2008). Implicit concerns about justice, however, form a silent frame in numerous grounded theory studies. Many researchers’ studies assume the significance of social justice goals (see, for example, Carter, 2003; Ciambrone, 2007; Hyde and Kammerer, 2009; Jones, 2003; Karabanow, 2008; Mcintyre, 2002; Roxas, 2008; Scott, 2005; Scott, London, and Gross, 2007; Wasserman and Clair, 2010) and other studies advance these goals through the content of their analyses (Frohman, 1998; Quint, 1965; Sakamoto et al., 2009; Valdez and Flores, 2005; Sixsmith, 1999; Swahnberg, Thapar-Björkert, and Berterö, 2007; Ullman, and Townsend, 2008; Veale and Stavrou; 2007). In keeping with grounded theory logic, social justice issues may arise through grappling with data analysis as well as through learning what is happening during data collection or starting from an explicit standpoint of pursuing social justice.
Researchers who pursue social justice goals enrich the contributions of development of the grounded theory method. Their attentiveness to context, constraint, power, and inequality advances attending to structural, temporal, and situational contexts in qualitative research generally and in grounded theory studies specifically. Social justice researchers are attuned to the silent workings of structure and power. They can offer grounded theorists important reminders of how historical conditions and larger social conditions shape current situations.
The critical stance of social justice inquiry combined with its structural focus can aid grounded theorists to locate subjective and collective experience in larger structures and increase understanding of how these structures work (Charmaz, 2005; Clarke, 2003, 2004; Maines, 2001; Rivera, 2008). The narrow focus and small size of many grounded theory studies have militated against the authors finding variation in their data much less seeing how structure and historical process affect both the data and analysis. Like most qualitative researchers over the past fifty years, grounded theorists have often concentrated on overt processes and overt statements. A social justice standpoint brings critical inquiry to covert processes and invisible structures. Thus, we can discover contradictions between rhetoric and realities, ends and means, and goals and outcomes. This stance furthers understandings of the tacit, the liminal, and the marginal that otherwise might remain unseen and ignored, such as latent sources of conflict. The critical edge of social justice inquiry can help us subject our data to new tests and create new connections in our theories (Charmaz, 2005).
Recent grounded theory studies show increased engagement with social justice issues. To varying degrees, these studies address power, agency, structural constraints, resources, and analyze a wide range of questions including specific problems of impoverished, oppressed, stigmatized, and disenfranchised people (Choi and Holroyd , 2007; Ciambrone, 2007; Hyde and Kammerer, 2009; Mevorach, 2008; Ryder, 2006; Sixsmith, 1999; Scott, London, and Gross, 2007; Tuason, 2008; Ullman and Townsend, 2008; Veale and Stavrou, 2007; Wilson and Luker, 2006; Wolkomir, 2001) as well as those that interrogate relationships between a social justice issue and social structure (Gunter, 2005; McDermott, 2007; Mitchell and McCusker, 2008). To date, the latter frequently emerge in the implications of studying a pressing issue or small group of people who suffer multiple and cumulative effects of iniquities (e.g. Dixon, 2007; Jiménez, 2008; Wasserman and Clair, forthcoming; Wolkomir, 2001, 2006; Valadez, 2008; Zieghan and Hinchman, 1999). Researchers in diverse disciplines and professions have primarily used grounded theory methods for small studies of individual behavior, as have researchers using other qualitative approaches. That does not, however, preclude adopting grounded theory to develop organizational and structural studies as studies in organizations (O’Connor, Rice, Peters, and Veryzer, 2003; Scott, 2005; Vandenburgh, 2001) and the sociology of science have already shown (Casper, 1998; Clarke, 1998; Star, 1989). Grounded theory methods have gained a foothold in participatory action research (PAR) (Dick, 2007; Foster-Fishman, Nowell, Deacon, Nievar, and McCann, 2005; Kemmis and McTaggart, 2005; McIntyre, 2002; Poonamallee, 2009; Sakamoto et al., 2009; Teram, Schachter, and Stalker, 2005), a method which holds powerful potential for re-envisioning life and thus for advancing emancipatory change.
Reconstructing Grounded Theory
Grounded Theory as a Specific, General, and Generalized Method
Grounded theory is simultaneously a method that invokes specific strategies, a general method with guidelines that has informed qualitative inquiry, and a method whose strategies have become generalized, reconstructed, and contested. Strauss and Corbin (1994) observed over fifteen years ago that grounded theory has become a general qualitative method. Grounded theory methodological strategies of simultaneous data collection and analysis, inductive coding, and memo-writing have permeated qualitative research. Authors who claim to use grounded theory may, however, be conducting a more general form of qualitative research. Some authors’ claims to be using grounded theory are attempts to legitimize inductive qualitative research; others result from naïve readings of the method. The abstract guidelines and dense writing in the early grounded theory texts led to misunderstandings of the method and confused readers (Piantanida, Tananis, and Grubs, 2004).
As grounded theory has become a general method, researchers may only adopt one or two grounded theory strategies (Foster-Fishman et al 2005; Mitakidou, Tressou, and Karagianni, 2008). Other researchers may adopt more strategies but misunderstand them. And consistent with Virginia Olesen’s (2007) statement about her work, some researchers may understand grounded theory strategies but their research questions and objectives lead them to combine grounded theory strategies with other qualitative approaches. Researchers frequently combine grounded theory strategies, especially coding with narrative and thematic analyses (see for example, Cohn, Dyson, Wessley, 2008; Hansen, Walters, and Baker, 2007; Harry, Sturges, and Klingner, 2005; Wilson and Luker, 2006, Mathieson and Stam, 1995; Moreno, 2008; Salander, 2002; Sakamoto et al, 2009; Somerville et al, 2008; Tuason, 2008; Williamson, 2006).
Naïve misunderstandings of grounded theory can prevent researchers from realizing its analytic power. In brief, misunderstandings about grounded theory arise in three main areas: coding, theoretical sampling, and theory construction. I outline these misunderstandings and describe principles of grounded theory coding here but discuss them more thoroughly elsewhere (Charmaz, 2006, 2007, 2008b, 2008c). Cathy Urquhart (2003) questions whether grounded theory is, in essence, a coding technique. Coding is crucial but grounded theory is much more than a coding technique. Many researchers, however, use it for just that and appear rely on CQDAS [Computer-assisted Qualitative Data Analysis Software] to do their coding.(see Bong, 2007 for problems of grounded theory coding with CAQDAS).
Grounded theory coding strategies include sorting, synthesizing, and summarizing data but, moreover, surpass these forms of data management. Rather, the fundamental characteristic of grounded theory coding involves taking data apart and defining how they are constituted. By asking what is happening in small segments of data and questioning what theoretical category each segment indicates, grounded theorists can take a fresh look at their data and create codes that lead to innovative analyses. By simultaneously raising questions about power and connections with larger social units, social justice researchers can show how data are constituted in ways that elude most grounded theorists.
Early grounded theory works (Glaser and Strauss, 1967; Glaser, 1978) lack clarity on what theoretical sampling means and how to conduct it. This lack of clarity combined with researchers’ preconceptions of the term “sampling” created frequent misunderstandings. Theoretical sampling occurs after the initial data collection and analysis. It means sampling data to fill out the properties of an emergent conceptual category (Charmaz, 2006; Glaser, 1978, 1998; Morse, 2007). This strategy also helps a researcher to discover variation in the category and differences between categories. Thus, grounded theorists conduct theoretical sampling after they have developed tentative categories of data, not before they begin to collect data.
The objective of theoretical sampling is theory construction. Jane Hood (2007) contends that textbook authors often mistake theoretical sampling for purposive sampling, which sets criteria for representation of key attributes when planning initial data collection. Sharon Nepstad’s (2007) methodological statement assumes this common misunderstanding: “Then I contacted staff members at these solidarity organizations and with their input I constructed a purposive theoretical sample (Glaser and Strauss, 1967) to ensure a diverse representation of geographic regions, age range, gender, and levels of participation in the movement” (p. 474).
In another area of common misunderstanding, many grounded theorists claim to construct theory but neglect to explicate what they assume theory encompasses. As I (Charmaz, 2006, p. 133) have argued, their assumptions about what constitutes theory suggest a range of meanings that include: 1) a description, 2) an empirical generalization, 3) relationships between variables, and 4) an abstract understanding of relationships between concepts. If we define theory as either explaining the relationships between concepts or offering an abstract understanding of them, most studies that purport to have produced theory actually do not. Their authors assert that they construct theory but their analyses attest to their efforts to synthesize data and condense themes. Despite lofty claims to the contrary, most grounded theorists do not produce theory, although some move toward theory construction. And numerous authors produce mundane descriptions under the guise of doing grounded theory. The potential of grounded theory for theory construction has yet to be fully explored and exploited.
Grounded Theory as a Specific Method
We can discern convergent approaches between recognized grounded theorists that undergird grounded theory as a specific method. How grounded theorists use their methodological strategies differs from other qualitative researchers who study topics and structures instead of actions and processes. How we collect data and what we do with it matters. Research actions distinguish grounded theory from other types of qualitative inquiry (Charmaz, In Press). Grounded theorists representing each version engage in the following actions:
1. Conduct data collection and analysis simultaneously in an iterative process
2. Analyze actions and processes rather than themes and structure
3. Use comparative methods
4. Draw on data (e.g. narratives and descriptions) in service of developing new conceptual categories
5. Develop inductive categories through systematic data analysis
6. Emphasize theory construction rather than description or application of current theories
7. Engage in theoretical sampling
8. Search for variation in the studied categories or process
9. Pursue developing a category rather than covering a specific empirical topic. (Charmaz, In press)
Researchers who engage in the first five actions give their studies a distinctive analytic cast that differs from than other qualitative works, particularly those that remain descriptive. Studies by grounded theorists reach across individuals and events to reveal a collective analytic story. Detailing conceptual categories takes precedence over participants’ accounts and summarized data. A grounded theorist presents excerpts and summaries of data to demonstrate the connection between data and category and to offer evidence for the robustness of the category. A much smaller number of researchers engage in the remaining actions but they move their analyses into theory construction (Charmaz, In press).
Despite agreement about these nine research actions, what stands as a bona fide grounded theory study may remain ambiguous (Charmaz, 2008e, In press; Timmermans and Tavory, 2007). Methods statements in published works seldom address analytic strategies, much less detail them. Certain studies such as Qin and Lykes (2006), Roschelle and Kaufman (2004) and Wolkomir (2001) illustrate distinctive grounded theory logic because they conceptualize a problematic process, construct analytic categories from inductive, comparative coding of data, define the properties of the categories, specify the relationships between categories, and outline the consequences of the processes.
If readers cannot discern a distinctive grounded theory logic in the analysis, then it becomes difficult to determine whether authors’ claims to using grounded theory methods are mistaken or are aimed to legitimize inductive qualitative research. Nonetheless, some authors’ analyses may not indicate a grounded theory approach but their methodological descriptions reveal a sophisticated understanding of the method. Consider Henry Vandenburgh’s (2001) statement in his study of organizational deviance:
` I followed the stages suggested by Turner (1981) in interpreting Strauss,
first developing categories that used available data to suggest nominal
classifications fitting these data closely. I then saturated these categories
by accumulating all of the examples I could from my interview data
that fit each category. Next, I then abstracted a definition for each category
by stating the criteria for putting further instances of this specific
type of phenomena into the category. I continued to use the categories
by making follow-up calls based upon some of the questions raised. I
then further exploited the categories by inspecting them to see if they
suggested additional categories, suggested more general or specific
instances, or suggested their opposites. I noted and developed links
between categories by becoming aware of the patterned relationships
between them, and by developing hypothesis about these links. Finally,
I considered the conditions under which the links held by theorizing
about these relationships and the contexts that conditioned them. I then made conditions to existing theory. (2001, p. 62)
Like Vanderburgh, other grounded theorists may reveal their use of the method in their methodological discussions rather than in their analyses. Monica Casper (1997; 2007) and Robert Thornberg (Thornberg and Charmaz, forthcoming) each have telling discussions that illuminate their studies (see Casper, 1998; Thornberg, 2007, 2009)
Grounded Theory as Contested from Within
Grounded theory is a contested method from both within and without. The contested status of the method further complicates what stands as a grounded theory study today. Since its inception in 1967, the grounded theory method has undergone both clarification and change by all of its major proponents. Grounded theory has become an evolving general qualitative method with three versions: constructivist, objectivist, and post-positivist. Major texts that teach readers how to use grounded theory represent each version of grounded theory (Bryant and Charmaz, 2007; Charmaz, 2006; Corbin and Strauss, 2007; Glaser, 1978; 1998; Strauss and Corbin, 1990, 1998).
Constructivist grounded theory adopts the methodological strategies of Glaser and Strauss’s classic statement but integrates relativity and reflexivity throughout the research process. As such, this approach loosens grounded theory from its positivist, objectivist roots and brings the researcher’s roles and actions into view. Constructivist grounded theory uses methodological strategies developed by Barney Glaser, the spokesperson for objectivist grounded theory, yet builds on the social constructionism inherent in Anselm Strauss’s symbolic interactionist perspective Charmaz, 2006; 2007, 2009a). Constructivist grounded theory views knowledge as located in time, space, and situation and takes into account the researcher’s construction of emergent concepts.
Objectivist grounded theory shares an emphasis on constructing emergent concepts but emphasizes positivist empiricism with researcher neutrality while aiming for abstract generalizations independent of time, place, and specific people (Glaser 1978, 1998, 2007). Unlike many positivists of the past, however, Glaser evinces little concern for establishing criteria for data collection or for evaluating its quality. He maintains that “all is data” (2001, p. 145) but leaves unexamined what researchers may define as “all.” For Glaser, a concern with data reflects the “worrisome accuracy” (Glaser, 2002, paragraph 2) characterizing the conventional qualitative research that he argues against. Phyllis Noerager Stern (2007), a major proponent of Glaserian grounded theory, finds that a small number of cases is sufficient to saturate the researcher’s emerging analytic categories. Glaser contends that examining many cases through the comparative process renders data objective. His earlier view that research participants will tell researchers their main concern about what’s happening in their setting (Glaser, 1992) likely contributed to the notion of discovering theory in the data, as though it simply resided there. I have long argued that we cannot assume that participants’ overt statements represent the most significant data (Charmaz, 1990, 1995, 2000, 2001). Instead their statements may take for granted fundamental processes that shape their lives or provide a strategic rhetoric to manage an impression (Charmaz, 1990, 2000, 2008f).
Constructivist grounded theory contrasts with its objectivist predecessor in several fundamental ways, as I indicate above and summarize below. Post-positivist grounded theory (Corbin and Strauss, 2008; Strauss and Corbin, 1990, 1998) takes a middle ground between the two versions. It places less emphasis on emergence than the objectivist and constructivist approaches, as it provides pre-conceived coding and analytic frameworks to apply to data. Yet post-positivist grounded theory views reality as fluid, evolving, and open to change. Strauss and Corbin’s early books made grounded theory a method of application rather than innovation (Charmaz, 2007).
In her recent reflection, however, Juliet Corbin (2009) outlines how her approach to research has changed. She describes having been imbued with methodological prescriptions of earlier decades that shaped writing the first two editions of Basics of Qualitative Research (1990, 1998). These prescriptions led qualitative researchers to: (1) study data to find the theory embedded in them; (2) maintain objectivity; (3) avoid “going native”; and (4) capture a semblance of “reality” in data and present it as “theoretical findings”, while simultaneously believing that no one truth existed (pp. 36-37). Corbin’s list combined with the technical procedures in Strauss and Corbin’s (1990, 1998) Basics of Qualitative Research confirms my earlier contentions (Charmaz, 2000, 2002) that their earlier editions contain objectivist threads. Corbin (2009) now, however, endorses engaging in reflexivity, takes a value stance that furthers social justice, believes in multiple realities, and disavows rigid application of technical procedures. These changes mark the updated third edition of Basics of Qualitative Research (Corbin and Strauss, 2008) and bring it closer to constructivist grounded theory.
The three versions of grounded theory share commitments to conceptualizing qualitative data through analyzing these data, constructing theoretical analyses, and adopting key grounded theory strategies. Each version emphasizes systematic inquiry using transparent strategies, begins with an inductive logic, emphasizes constructing theory, and aims to construct useful analyses for research participants, policy-makers, and relevant practitioners (Charmaz, 2009b). Which strategies each version adopts, creates, or discards often differ in crucial ways that reflect more than favored or disfavored techniques. Differences in epistemology and ontology come into play.
Epistemological Differences in Versions of Grounded Theory
Grounded theory contained the seeds of divergence from its beginnings. Glaser’s Columbia University positivism and theoretical background in structural-functionalism and Strauss’s University of Chicago pragmatism drew on conflicting philosophical and methodological presuppositions about the nature of reality, objectives of inquiry, and the research process and practice. The legacy of Anselm Strauss rests on pragmatism and its development in symbolic interactionism (Charmaz, 2008d). Differences between a grounded theory informed by positivism and one informed by pragmatism appear most starkly in the contrast between objectivist grounded theory and constructivist grounded theory articulated in the second edition of this Handbook (Charmaz, 2000; see also Bryant, 2002; Bryant and Charmaz, 2007, In press; Charmaz, 2002, 2006, 2007, 2008e, 2009a, 2009c; Charmaz and Bryant, In press; Charmaz and Henwood, 2008).
Objectivist grounded theory assumes that a neutral observer discovers data in a unitary external world. In this view, researchers can separate their values from “facts” residing in this world and suggests what Kelle (2007) calls “epistemological fundamentalism” (p. 205). In this approach, data gathering does not raise questions about researchers’ tacit assumptions, privileged statuses, or the particular locations from which they view studied life. The researcher stands outside the studied phenomenon. Data are “there” rather than constructed. Researchers can add reflexivity about data collection and their roles, if they wish. Ordinarily, however, the neutral but passive observer simply gathers data to analyze as the authoritative expert and active analyst. In the objectivist logic, the number of cases corrects the researcher’s possible biases. This approach gives priority to the researcher’s voice and analysis and treats the researcher’s representation of participants as straightforward, not as inherently problematic. A hazard is that researchers may import their unacknowledged presuppositions into the research process and product. Objectivist grounded theory aims for parsimonious abstract generalizations about relationships between variables that explain empirical phenomena. These generalizations constitute a middle-range theory explaining the studied phenomenon.
Constructivist grounded theory adopts a contrasting relativist approach that shifts its ontological and epistemological grounds (Charmaz, 2009c) and aligns them with the pragmatist tradition of Anselm Strauss (see Charmaz, 2009a; Reichert, 2007; Strübing, 2007). Here, realities are multiple and the viewer is part of what is viewed. Subjectivities matter. Values shape what stands as fact. To the extent possible, constructivist grounded theorists enter the studied phenomenon and attempt to see it from the inside. Researchers and participants co-construct the data through interaction. Data reflect their historical, social, and situational locations, including those of the researcher. Representations of the data are inherently problematic and partial. These concerns involve constructivist grounded theorists in reflexivity throughout inquiry as an integral part of the research process (see also Mruck and Mey, 2007; Neill, 2006). Rather than aiming for theoretical generalizations, constructivist grounded theory aims for interpretive understanding. The quest for generalizations erases difference and obscures variation (see also Clarke, 2003, 2005, 2006; Clarke and Friese, 2007). For constructivists, generalizations remain partial, conditional, and situated. Moreover, generalizations are not neutral. As Norman Denzin (2007) avows, interpretation is inherently political.
All these contrasts alter the processes and products of inquiry, as do differences in grounded theory practice, such as the contested place of the literature review. Glaser (1978, 1998, 2003) advocates conducting the literature review after developing an independent analysis to avoid forcing the data into preconceived categories and theories. However, few doctoral students and professional researchers begin their studies without knowledge of their fields (Charmaz, 2006; Lempert, 2007). They must include thorough literature reviews in dissertation proposals, grant applications, and today even in some human subjects IRB [Institutional Review Board] applications. Karen Henwood and Nick Pidgeon’s (2003) concept of theoretical agnosticism makes more sense than theoretical innocence. They argue that researchers need to subject all possible theoretical explanations of a phenomenon to rigorous scrutiny---whether from the literature or their own analysis. Perhaps most significantly, constructivist grounded theorists contend that researchers’ starting points and standpoints, including those occurring throughout inquiry, influence the research process and product.
Grounded Theory in Mixed Methods Social Justice Inquiry
Researchers have identified grounded theory as a useful qualitative method to adopt in mixed methods research. Despite the growing number of studies that purport to use grounded theory studies in mixed methods complete knowledge of the topic projects, few of these studies have a clear focus on social justice. The place of grounded theory in mixed methods social justice inquiry has yet to be developed. Thus, I offer brief concerns about mixed methods here that grounded theory social justice researchers might consider.
Mixed methods research usually means using both quantitative and qualitative methods to gain more and a more nuanced analysis of the research problem. Definitions of mixed methods and what they mean for inquiry are, however, contested and multiple. For mixed methods specialists (see, for example, Cameron, 2009; Creswell, 2003; Morgan, 2007) the rapid rise of mixed methods is a movement heralding a paradigm shift analogous to the qualitative revolution that Denzin and Lincoln proclaimed in 1994. For many researchers, mixed methods are tools that produce findings, regardless of their reasons for adopting these tools, analyzing the findings, and deciding whether and to what extent to use each of the subsequent analyses. For a few researchers, mixed methods simply means using more than one method, whether or not these methods mix quantitative and qualitative research. R. Burke Johnson, Anthony J. Onwuegbuzie and Lisa A. Turner (2007) view mixed methods as combining qualitative and quantitative approaches including their respective perspectives, analyses, and forms of inference. They point out that mixed methods mean combining elements of methods for “breadth and depth of understanding and corroboration” (p. 123).
The discussion of mixed methods takes into account Norman Denzin’s (1970) early call for triangulation (Greene, 2006; Morse, 1991; Tashakkori and Teddlie, 2003). Numerous researchers advocate methodological pluralism. Others take a more skeptical view and see the quantitative data and analysis as not only dominating mix methods projects, but also “quantitizing” qualitative data by transforming them into numbers (Sandelowski, Voils, and Knafl, 92009). However, Creswell, Shope, Plano Clark, and Green (2006) and Creswell and Plano Clark (2007) argue that qualitative methods extend mixed methods practice and may be given priority in mixed methods projects. In practice, researchers use mixed methods for varied purposes including to: (1) construct instruments, (2) corroborate findings, (3) reduce cultural and investigator biases, (4) improve clinical trials, (5) address research participants’ experience, (6) demonstrate credibility, (7) increase generalizability, and (8) inform professional practice and/or public policy.
Questions arise about integrating the results and analyses in mixed methods studies. To what extent should integration of quantitative and qualitative findings be a major methodological goal? What should be done when the qualitative and quantitative data have conflicting results? Bryman (2007) finds that mixed methods researchers often dismiss the qualitative analyses. He contends that “the key issue is whether in a mixed methods project, the end product is more than the sum of the individual quantitative and qualitative parts” (p. 8). In practice, that may not occur, nor may it have been the researchers’ intent, as Bryman observes.
Mixed methods research designs often consist of complicated procedures and hence require team efforts. Grounded theory mixed methods projects are steadily increasing in fields such as education and health in which funded team research is common. Social justice research, especially in its explicit forms, is less likely to be a funded team project staffed by an array of methodological specialists having different but complementary skills. Few researchers are equally skilled in both quantitative and qualitative methods. Social justice research is likely to be an unfunded individual pursuit or a participatory action research project in which the researchers are members of and responsible to local communities.
Christ (2009) correctly observes that the goals of mixed methods research typically contrast with those in transformative inquiry in which social justice goals dominate. As he points out, researchers conduct critical and transformative research “to improve communities or reduce oppression, not to generalize results from a non-representative sample to a larger population” (p.293). However, Donna Mertens (2007, 2010) argues eloquently for using mixed methods in a transformative paradigm to further social justice and Deborah K. Padgett (2009) states, “Social justice values do not have to be sidelined” (p. 101). Their purposes are explicit rather than hidden under a bland—and exclusive—term like “public sociology” (Burawoy, 2004).
Because social justice researchers may face skeptical audiences, presenting multiple forms of data in an integrated analysis may buttress their reports. The emerging philosophical foundations for mixed methods would support their efforts. Discussions are occurring that position mixed methods in pragmatism, and thus fit grounded theory research in social justice (see, for example, Duemer and Zebedi, 2009; Feilzer, 2010; Morgan 2007).
Researchers in education are among the most attune to using grounded theory in mixed methods studies for social justice goals. As is evident in other studies, however, other researchers may assume rather than state social justice goals and use grounded theory in limited or extensive ways. To cite one interesting example, Sahin-Hodoglugil, vander Straten, Cheng, Montgomery, Kcanek, Mtetewa, Morar, Munyaro, Padian, and the MIRA Team (2009) used mixed methods in a randomized controlled clinical trial to study the effect of low-cost HIV prevention method, the diaphragm. This method gave women control because they could use it without their male partners’ knowledge. The authors invoked an iterative process in which both quantitative data and qualitative findings informed each other. Sahin-Hodoglugil et al used insights about covert diaphragm use from the qualitative analysis to inform the analytic framework for the quantitative data and some findings in the quantitative portion were subsequently explored by gathering qualitative data. These researchers discovered that covert use was more complicated than they had anticipated and occurred along a continuum with disclosure.
In short, social justice researchers who can bring multiple types of solid data to their analyses make their reports less easy to dismiss. The test of mixed methods studies resides in doing credible work in all methods adopted to answer the research questions, fulfill the research goals, and convince relevant audiences of the significance of the reports.
Using Grounded Theory Strategies in Social justice Inquiry
In this section, I offer several specific examples of grounded theory in practice. Coding, memo-writing, theoretical sampling and saturation, sorting memos are all part of the process. These grounded theory strategies have been described elsewhere in detail (Charmaz 1983, 1990, 2001, 2002, 2005, 2006; Corbin and Strauss, 2008, Glaser, 1978, 1998, 2001, 2003, Strauss, 1987; Strauss and Corbin, 1990, 1998) so I merely introduce several examples of how constructing grounded theory analyses animate social justice inquiry.
Coding for Processes
By using gerunds to code for actions, grounded theorists make individual or collective action and process visible and tangible. Social justice researchers can use grounded theory coding strategies to show how people enact injustice and inequity. Gerunds define actions and enable grounded theorists to envision implicit actions and to identify how they are linked (see, for example, Schwalbe, 2005; Schwalbe, Goodwin, Schrock, Thompson, and Wolkomir, 2000).
Coding data for actions and mining the theoretical potential of both data and codes make grounded theory distinctive (Charmaz, 2006, 2008b, 2008c). Coding with gerunds pinpoints actions and thus helps grounded theorists to define what is happening in a fragment of data or a description of an incident. Gerunds enable grounded theorists to see implicit processes, to make connections between codes, and to keep their analyses active and emergent. In contrast, coding for topics and themes helps the researcher to sort and synthesize the data but neither breaks it apart as readily as grounded theory coding for actions nor fosters seeing implicit relationships between topics and themes.
Line-by-line coding, the initial grounded theory coding with gerunds, is a heuristic device to bring the researcher into the data, interact with it, and study each fragment of it (see Textbox 1). This type of coding helps to define implicit meanings and actions, gives researchers directions to explore, spurs making comparisons between data, and suggests emergent links between processes in the data to pursue and check. The data excerpts in the textboxes tell the story of a middle-aged woman with lupus erythematosus whose friends rushed her to a doctor to reassess her medications during a medical crisis. These medications often cause multiple side-effects including confusion, depression, blurred vision, and inappropriate emotional responses. After two hospital transfers, this woman’s medical crisis became redefined as a psychiatric crisis. Subsequently, her claims of having physical symptoms were unacknowledged and her requests for lupus medications went unheeded. She aroused the doctor’s ire when he discovered another patient helping her complete the detailed confidential in-take survey. Getting help with the survey broke hospital rules but this woman’s vision problems meant that she could not read the survey and so just filled circles randomly after her doctor forbade her from having help. Her actions in one incident after another made sense given her situation, but fit neither the hospital protocol nor her treatment program. Nevertheless, she attempted to present her views and to become her own advocate while her illness worsened. A psychiatrist who is unaware of her medical history or ignores it could invoke the same incidents as justifying his treatment approach. In this case, the grounded theory codes chronicle the woman’s progressive loss of control over her life and her illness. Thus the initial codes in the excerpt become the details substantiating a more general code, “resisting spiraling powerlessness.”
Place Textboxes 1 and 2 here side by side, if possible.
Comparing the codes in the textboxes demonstrates that grounded theory codes preserve the character of the data, provide a precise handle on the material, and point to places that need further elucidation. Coding gives the researcher leads to pursue in subsequent data collection. In vivo codes use research participants’ terms as codes to uncover their meanings and understand their emergent actions. Zieghan and Hinchman’s (1999) in vivo codes: “breaking the ice,” “figuring out how to help,” “trying to understand” gave form to their study of college students who tutored adult learners. Note that they code in gerunds and thus portray the tutors’ actions as they wrestled with dealing with their situations. Despite the student tutors increased awareness of poverty and lack of opportunity, the authors learned “that the border between campus life and the adult literacy community is a site of reproduction rather than transformation” (p. 99).
If coding in gerunds is so fruitful, why don’t more researchers use them? In my view, the English language favors thinking in structures, topics, and themes rather than thinking in actions and processes. In addition, Strauss and Corbin’s (1990, 1998) books have instructed thousands of researchers but emphasize gerunds less than Charmaz’s (1990, 2006, 2008c) and Glaser’s (1978, 1998) works. Many researchers report beginning with open coding, the initial coding in which the researcher examines and categorizes the data. Some turn next to axial coding, a type of coding to relate categories to subcategories, or to thematic coding but do not build fresh conceptual categories. Those following Strauss and Corbin (1990, 1998) often adopt complicated coding procedures to generate themes (Ball et al, 2009; Morrow and Smith, 1995; Sakamoto et al, 2009; Ullman and Townsend, 2008). Ullman and Townsend’s (2008) coding procedures generated themes such as “Definitions of Feminist/Empowerment Approaches,” “Importance of Control,” “Techniques for Empowerment,” and “Advocate Versus Agency Orientations,” rather than a theory or conceptualized process. Although many authors found Strauss and Corbin’s coding procedures to be helpful, some like Judy Kendall (1999) did not. She states, “I became so distracted by working the model to its natural conclusion that I stopped thinking about what the data were telling me in regard to the research question” (p. 753).
Grounded theory coding need not be complex. By engaging in thorough coding early in the research process and comparing data and codes, the researcher can identify which codes to explore as tentative categories. In turn, selecting categories expedites inquiry because the researcher then uses these categories to sort large batches of data. This approach is particularly useful in social justice research projects that address pressing social issues and policies. Grounded theory coding preserves empirical detail and simultaneously moves the project toward completion.
Increasingly, grounded theorists turn to one of the CAQDAS programs for coding the data. Several CAQDAS programs aim to be compatible with grounded theory logic and treat qualitative inquiry as interchangeable with grounded theory, or their conception of grounded theory. Software developers may have been criticized for their reliance on grounded theory. Ironically, however, their products may fit general qualitative coding for topics and themes more than coding for processes and engaging in comparative analysis. Grounded theory coding involves more than merely applying labels, identifying topics, and labeling themes, although many researchers do not realize it.
Depending on the grounded theorist’s skills and objectives, advantages of using CAQDAS may: (1) include relative ease of searching, retrieving, sorting, separating, and categorizing data and codes, (2) the ability to work at multiple levels of analysis simultaneously, (3) visibility of both the data and analytic processes (4) document-sharing capacities for team research, and (5) management and organization of the data and emerging analysis . Since the advent of CAQDAS, numerous researchers (see, for example, Fielding and Lee, 1998; Glaser, 2003; Weitzman, 2000) have raised varied questions concerning the conceivable advantages of using CAQDAS. Their concerns included users becoming too close or too distanced from their data, software design driving the analysis, and users being able to produce results without understanding the analytic process or range of analytic approaches. CAQDAS has gained much wider audiences since the early years and the software supports more intricate functions (Fielding and Lee, 2002). As the software becomes more sophisticated, its effects on knowledge-production may change. Bringer, Johnston, and Brackenridge (2006) state that effective use of the grounded theory method holds greater significance than whether or not a software package is adopted. Konopásek (2008) avers that “The software… extends the researcher's mental capabilities to organise, to remember, and to be systematic. But while doing so it essentially remains a stupid instrument” . Yet for Konopásek, the software not only consists of tools but also a virtual environment in which a set of mediations and embodied and practice-based knowledge production occurs.
Udo Kelle (2004) argues that CAQDAS requires its users to explicate their data management strategies and subsequently to think about their methodological and epistemological significance (p. 473). Does it? To what extent? Kelle acknowledges that CAQDAS could be viewed as a step in the rationalization and mechanization of qualitative research with the production of trivial results. The plethora of simplistic CAQDAS papers written under the guise of grounded theory confirms that this problem exists. However, Kelle contends that CADQAS helps users to gain clarity about the processes involved in theory construction. Yet many researchers use CADQAS—and grounded theory—primarily for coding without venturing into theory construction. Nonetheless, Kelle presents a worthy goal. But first, researchers need to learn to use the method.
Defining Extant Concepts through Data Analysis
Social justice researchers use concepts that reflect structural arrangements and collective forces. They could adopt grounded theory strategies to refine or redefine these concepts by their empirical properties. In this sense, they can subject a sensitizing concept to rigorous empirical analysis. In the following example, I examined my data about the experience of illness and used marginalization as a sensitizing concept (Blumer, 1969; van den Hoonard, 1997) from which to begin analysis. A 46-year old woman I call Marilyn (Charmaz, 2008f) recounted becoming ill and disabled with chronic fatigue syndrome and environmental illness.
Marilyn compares and contrasts her now unending saga of illness with the story of her past successes.
I did a lot of things that were very challenging, and, you know, I used to work 50, 60 hours a week, and made good money and had great benefits, and had a life and all of that stuff changed one year—really abruptly in one year. And since then, everything is gone—from the financial to memory to—and everything in between. (p. 7)
….Marilyn planned to testify at a city council meeting to restrict wood burning because of its devastating effects on people with lung disease, asthma, and chemical sensitivities. Because she anticipated possible discomfort from odors in a packed, closed room, she arrived 15 minutes before the time for community members to speak, but she had a long wait. Marilyn recounted:
I had to wear my mask and by two hours the charcoal [filter in the mask] was shot and by the time I got up to speak, you know, my voice was going, my brain was starting to go, I was having problems formulating words, so it’s, and then of course when you wear a mask, people, you know, you’ve seen mothers kind of pull their kids close to them. (pp. 8-9)
In her first statement, Marilyn is making identity claims about who she had been. Her second statement reveals the combined effects of appearance and of losing control over timing. The subsequent increasing visibility of Marilyn’s difference led to her being discredited, devalued—othered. As I coded data and compared incidents, it was apparent that visible difference marginalized people with illnesses and disabilities. But what did marginalization mean? Through coding data, I identified properties of marginalization that linked its social origins with subjective experience and thus stated, “Marginalization means boundaries or barriers, distance or separation, and division or difference. Disconnection, devaluation, discrimination, and deprivation exemplify experiences of marginalization” (p. 9). Moreover, I showed how people enacted marginalization.
Using grounded theory strategies in similar ways can assist social justice researchers to infuse taken-for- granted concepts with specific meanings. Furthermore, it can help them avoid reifying, objectifying, and universalizing ideas without putting them to test.
Developing Categories and Discovering Meanings
Grounded theory has long been touted as a method of discovery—of data and of theory. The constructivist critique argues that such “discoveries” are constructions located in space, time, and situation. Yet grounded theory can give us tools for constructing new understandings. Learning how participants define their situations, attempting to grasp what they assume, and understanding the problems that confront them become major sources of our “discoveries” and of the categories through which we conceptualize these discoveries. In his study of street youth, Jeff Karabanow (2008) discovered what leaving the street meant to these young people. He explains a crucial category and part of the process of leaving the streets as “cutting street ties” (p. 781).
Cutting street ties meant leaving friends, surrogate families, and a culture associated with the downtown core. For many young people, friends and surrogate families were forged as a result of, or during, very stressful survival situations.
Karabanow’s category speaks to meanings of the past as well as of the present and future. Cutting street ties meant more than merely leaving the streets. It occurred in a complex context in which the streets often held greater appeal to the young people than having shelter. Karabanow’s analysis has resonance and power not only because of the clarity of his analysis but also because of the strength of his data: 128 interviews with street youth, 50 interviews with service providers, lengthy experience with the topic, and the help of two research assistants who lived on the streets.
Note that Karabanow’s category, cutting street ties is a precondition for the larger process of leaving the streets. Grounded theory analyses gain this kind of specificity when researchers scrutinize the data to explicate the conditions that produce the studied process or phenomenon. In their study of women who had suffered childhood sexual abuse, Susan L. Morrow and Mary Lee Smith (1995) looked for causal conditions that led to the abuse. They identify two strategies that their research participants had used to cope with it: “keeping from being overwhelmed by threatening or dangerous feelings,” and “managing helplessness, powerlessness, and lack of control” (p. 27-28). Morrow and Smith find that these women had had few resources available for help and thus had adopted psychological strategies that focused inward on self and emotions such as reducing the intensity of troubling feelings, avoiding or escaping feelings, or using self-induced physical pain to override emotional pain.
As Morrow and Smith learned what these women had done to cope with their situations, they also learned the meanings the women had held. Social justice inquiry often focuses on people who experience horrendous coercion and oppression. Not surprisingly their categories reflect the untenable situations they observe. Angela Veale and Aki Stavrou (2007) studied the reintegration of Ugandan child abductees who had been forced to fight against the Uganda People’s Defense Force (UPDF), their own people. Veale and Stavrou state that the child abductees are part rebel soldier, part a child of his or her village but yet identified as “external to it—as the aggressor” (p. 284). Veale and Stavrou’s category depicting this conflicted identity is “Managing Contradictions.” Veale and Stavrou state:
The UPDF is Ugandan, and fighting the UPDF is a source of sadness, because they fought to kill the enemy in order to survive. Victor expressed the conflict as follows:
Victor: When fighting against the Ugandan Army, I felt partly as army, partly as civilian.
Interviewer: You used to steal food from Ugandan families. What did you feel when you steal food?
Victor: I feel very bad ’cause the food I am going to steal is [from] my father or guardian, my brothers’ or sisters’ guardian.
For these youth, this dual role as soldier–abductee could not be resolved. (P. 285)
The irony of Veale and Stavrou’s category resides in the impossibility of resolution: the contradictions exceeded the confronted reality and yet these very contradictions were the reality.
Each of the categories discussed above remain close to the studied experience and address what is happening in the data. Grounded theorists construct theoretical categories, in contrast, as they ask what theoretical questions and concepts the data indicate and Michelle Wolkomir’s (2001) analysis suggests below.
Conceptualizing a Process
The emphasis on coding in action terms enables grounded theorists to discern processes that might otherwise remain invisible. Scrutinizing these processes can help social justice researchers refine their concepts, form nuanced analyses, see how powerful cultural scripts are acted upon, and become attune to possibilities for change. In her study of gay and ex-gay Christian men in a support group, Michelle Wolkomir (2001) outlines how the men engaged in “ideological maneuvering” (p. 407) to evade and subvert Christian ideology that condemned their sexuality and viewed them as “egregious sinners” (p.408). She argues that such ideological revision requires sustained effort, particularly when conducted by marginalized groups without power.
Consistent with a grounded theory emphasis on analyzing social and social psychological processes, Wolkomir’s major conceptual category, ideological maneuvering, is a process. She developed her analysis of this process through studying the men’s actions and observing the tensions they faced from their perspectives. How could they avoid stigma and claim moral Christian identities? Wolkomir’s guiding analytic question for her article asks, “Under what conditions is such change [ideological change] likely to occur, and how is it accomplished?” (p. 407). By raising such questions and defining these conditions, Wolkomir brings analytic precision to her analysis. Moreover, her work provides a theoretical concept that can be transported and tested in other empirical studies.
Wolkomir’s article reveals the underpinnings of her grounded theory analysis while simultaneously providing an insightful analysis of the overall process and major conceptual category. Wolkomir states that the process of ideological maneuvering entails three subprocesses: (1) “selective dismantling of existing ideology to open new interpretive space; (2) constructing a new affirming ideology; and (3) authenticating new self-meanings” (p. 408). She treats these subprocesses as analytic categories and then demonstrates the actions constituting each one. Note that Wolkomir’s categories are active, specific, and rooted in the data. Her categories depict how the men dealt with the Christian ideology that condemned and excluded them. Wolkomir found that for one support group, dismantling the existing ideology explicitly included “redefining sin” (p. 413). These men discovered new Scriptural reasons to believe that the significance of homosexual sin had been exaggerated and “concluded that their homosexual sin was no worse than selfishness or gossip” (p. 414).
Not only does Wolkomir show how these men challenged and shifted reigning ideas and hierarchical relationships but she also specifies the conditions under which changes occur. Wolkomir’s analysis does not end with successful ideological maneuvering. Instead she positions her analysis in relation to its larger implications of her study. Wolkomir concludes that inequalities limit such ideological revision and, in turn, ideological maneuvering reproduces inequality because it allows the larger oppressive ideology to remain intact. In short, Wolkomir’s grounded theory analysis advances our understanding of how ideological change can occur while simultaneously specifying its limits.
Wolkomir’s processual analysis demonstrates grounded theory in practice. Her approach reveals how people confer meaning on their situation and enact ideological stances. Yet Wolkomir’s analysis does more. It contains strong links between detailed ethnographic description, substantive processual categories, and development of a theoretical concept, ideological maneuvering. Wolkomir then situates her concept and frames her article in the larger theoretical discourse on ideology, and by doing so offers a dynamic analysis of relationships between agency and structure. Wolkomir’s nuanced theoretical account contributes to knowledge in a substantive area, theoretical ideas in her discipline, and useful understandings for social justice scholars and activists.
Defining variation in a process or phenomenon is an important grounded theory goal, particularly of the postpositivist and constructivist versions. Researchers who conduct thorough research may discover variation within their findings and subsequent analyses. Learning how to handle variation analytically and how to write about it strengthens the analytic precision and usefulness of the grounded theory.
Grounded theorists compare their analyses with the extant literature, which can serve as data to illuminate the properties of emergent categories. Ordinarily, grounded theorists develop their analyses first and then use the relevant literature for comparative analysis. Researchers who gain intimate knowledge of their participants and settings may, however, define sharp differences with the literature early in their research. They may then construct their analyses from this position. Both Wasserman and Clair (2010) and Roschelle and Kaufman (2004) discovered that homelessness was not monolithic and sought to demonstrate the variation they found. Roschelle and Kaufman focused on homeless kids in their ethnographic study of an organization that served and sheltered homeless families. They offered new representations of these children that challenged earlier conclusions of homeless children having developmental and psychiatric problems. In a similar logic, Wasserman and Clair argued that homeless men who had networks on the streets were safer than those who used shelters.
The features of the setting and context of the studied phenomenon thus shape behavior and events. Jackson-Jacobs (2004) realized that he had found a strategic site that challenged earlier knowledge about crack cocaine uses and their worlds. He analyzed the settings and context of crack cocaine use among four college students as a strategic site and made systematic comparisons between it and previously reported sites of crack use.
Following Glaser and Strauss’s (1967) guidelines, Jackson-Jacobs treated the literature as data to analyze the variation between his strategic site to arrive at causal generalizations and thus build theory. Jackson-Jacobs revealed that two conditions in his study contrasted with prior research on crack cocaine use and altered our knowledge of it. First, college student crack cocaine users could keep their drug use bounded because they (1) had resources, (2) wanted to avoid being identified as crack users, (3) treated smoking crack as a leisure pursuit, (4) purchased crack from friends, and (5) gave higher priority to their conventional involvements. Second, these students had substantial residential mobility within a “safe” area where college students lived, not where drug dealers and users hung out. The residential location gave the men a benign environment that camouflaged hard drug use. Mobility allowed them to move if tensions arose or when they feared being identified.
By following the men over time, Jackson-Jacobs witnessed the explanatory power of the two conditions he had specified. One man lost control of his drug use and suffered the stigma of his friends’ viewing him as a failure and of his mother discovering his crack habit. Another man’s situation changed upon moving home. Location matters. This man no longer had the safety and relative anonymity afforded by his former neighborhood. He and his suburban friends now had to buy rock cocaine in the nearby urban ghetto, which changed the conditions, meanings, and consequences of drug use. They got into trouble with the dealers and police and experienced violence at the hands of both.
Comparative Analysis with Extant Literatures
In the above ethnographies, the authors report that they were struck by the difference between the portrayals of their studied phenomenon in the literature and what they later observed. In conventional grounded theory practice, researchers develop their analyses first and then return to the literature, whether to position their studies or to use the literature as data. Roz Dixon (2007) delineates how she developed her analysis in her exploratory study of the early school experiences and peer relationships of 35 deaf adults. Early coding revealed that deaf children’s classmates subjected them to physical and psychological attacks. Dixon’s early codes consisted of incidents such as “pulled my hearing aids out,” “damaged hearing aids,” and “banged my ears.” The children not only rejected a deaf classmate, but also colluded so that this child broke group norms, as the following account indicates:
Bryony: If there was a lot of noise, I didn’t stand a hope of hearing
anything . . . and frequently when things were going on and (the
teachers) were trying to tell things, people would start drumming
desks (she demonstrates making the sound of very gentle knocking
on the table top).
Interviewer: So that you couldn’t hear?
Bryony: So that I couldn’t hear. … I wouldn’t know what (had been said)…and the teacher would get pissed off with having to keep repeating, you know, “How many times do I have to tell you” sort of thing, and you’d say “Sorry,
I didn’t hear”. . . “Well, it’s funny you heard for the last half hour.”. . .
and, you know, and I think the teachers were very suspicious of me.” (Dixon, 2007, p. 12)
Apparently, Bryony had had no troubles with friendships at school until classmates excluded her after she became deaf. Dixon does not specify how the teacher’s exasperated response contributed to the simultaneous processes of experiencing ostracism and harassment. The teacher likely gave her students license to break classroom deportment rules and further harass Bryony. Dixon’s detailed grounded theory coding of her interview data revealed signs of ostracism such as in Bryony’s interview. Subsequently, Dixon used the literature as data to identify general properties of ostracism, the properties of temporary coercive ostracism, and the properties of actual exclusion. She states:
It was during the process of subsequently organizing the codes that some codes seemed particularly suggestive of ostracism. To test this hypothesis, a literature review was conducted to clarify the nature, function and parameters of ostracism. A set of codes was developed describing behaviours and contextual factors which might be seen in the interview data if ostracism had been at work. All data was reanalyzed. (2007, 9)
Consistent with conventional grounded theory practice, Dixon first coded her data and studied her codes. Subsequently, she examined and coded how other authors treated ostracism. This coding enabled her to define types of ostracism as well as the “generic features of ostracism” (pp.13-14), which she tested with her data. While acknowledging inherent limitations of using retrospective accounts, Dixon distinguishes conditions that link ostracism and bullying and specifies conditions under which ostracism may be desirable. Dixon’s grounded theory analysis led her to define problems related to ostracism by children and creative interventions for handling it.
Summary and Conclusions
The implications of the above discussion are five-fold. First, the review of grounded clarifies strategies and approaches that grounded theorists share. These strategies and approaches distinguish the method from other types of qualitative research. Simultaneously, the influence of grounded theory on the development of qualitative methods becomes more apparent.
Second, delineating the similarities between versions of grounded theory and juxtaposing their differences creates a space for methodological explication of foundational assumptions and of research practice. Grounded theory studies range between objectivist and constructionist approaches and often contain elements of both. Yet attending to foundational assumptions and to research practice constitute pivotal turns toward engaging in reflexivity. And that may heighten awareness of our research choices and actions and, moreover, deeper understandings of our research participants’ situations for we see them in new light.
Third, the constructivist version reclaims grounded theory from being a method of application to a method of innovation. Wolkomir’s (2003) analysis exemplifies the difference between application and innovation. She uses the method to learn about her participants’ views and concerns, not to apply a set of rules on her data. Under these conditions, grounded theory remains an emergent method. Both the form and specific content of the method arise as the researcher grapples with the problems at hand. Thus, the emergent character of the method contributes to its flexibility. This flexibility gives social justice researchers a mutable frame for their studies that they can adapt to fit their research problems and budding analyses.
Fourth, constructivist grounded theory acknowledges the foundations of its production, calls for reproduction of the method on new grounds and moves inquiry beyond what is overt and obvious. In these ways, constructivist grounded theory answers earlier criticisms of objectivist grounded theory that emanate from feminist scholarship (Olesen, 2007), postmodernism, performance, and interpretive ethnography (Denzin, 2007) as well as critiques of qualitative methodologists. Constructivist grounded theory challenges positivist elements that ignore reflexivity, overlook ethical issues, disregard issues of representation and do not attend to researchers’ agency in constructing and interpreting data (Olesen, 2007).
Fifth, constructivist grounded theory acknowledges the dual roots of the method in positivism and pragmatism and seeks to develop the emphasis on pragmatism. Consistent with pragmatism, constructivist grounded theory acknowledges multiple perspectives and multiple forms of knowledge. As such, its practitioners become attuned to nuances in empirical worlds that elude researchers who assume a unitary method and unitary knowledge, and are ill-quipped to grasp. These nuances include the unheard voice of dissent and the silence of suffering. Both may remain imperceptible when researchers use objectivist grounded theory or, for that matter, conventional research methods. Classic grounded theory set forth tools for developing theoretical sensitivity. Constructivist grounded theory adds tools for increasing critical sensitivity and thus holds considerable potential for social justice inquiry.
The constructivist turn in grounded theory has clarified the strategies of the classic statements and generated a resurgence of interest in the method. It complements new developments such as Adele Clarke’s (2005) situational analysis and incorporates methodological developments. It offers mixed methods researchers a set of useful tools and holds promise of informing software development. Constructive grounded theory is and will be a method for the twenty-first century.
 Such emphases often start with pressing social problems, collective concerns, and impassioned voices. In contrast, Rawls’ (1971) emphasis on fairness begins from a distanced position of theorizing individual rights and risks from the standpoint of the rational actor under hypothetical conditions. Conceptions of social justice must take into account both collective goods and individual rights and recognize that definitions of rationality as well as of “rational” actors are situated in time, space, and culture—and both can change. To foster justice, Nussbaum (2000:234) argues that promoting a collective good must not subordinate the ends of some individuals over others. She observes that women suffer when a collective good is promoted without taking into account the internal power and opportunity hierarchies within a group.
 Throughout my discussion, I focus on studies that identify grounded theory as their methods of inquiry.
 Their approach may follow conventions of framing research for academic consumption rather than indicating a stance on social justice.
 One of the originators of grounded theory, Barney Glaser has consistently argued that researchers can use the method for quantitative as well as qualitative research and his (2008) recent book reaffirms this argument. To date, few researchers have acted on it.
 Theoretical sampling means sampling to develop the properties of a theoretical category, no to sample for representation of a population.
 The task of offering methodological discussions has been taken up in other venues, such as this handbook. Publishers’ qualitative methods lists, journals such as Qualitative Inquiry and International Journal of Social Research Methods as well as methodological articles in substantive journals bring methodological proclivities and practices into view. Where once methodological confessional tales focused on what happened in the research site, now authors such as Wasserman and Clair (In press); Suddaby and Greenwood (2005), Harry, Sturges, and Klingner, (2005) reveal remarkable candor in making their analytic strategies transparent. These authors invert the backstage of analytic work and bring it to the frontstage of discussion. Although I view each discussion as only partly grounded theory (but see Suddaby’s  astute depiction of what is not grounded theory), I admire their candor and willingness to enter the methodological fray.
 I have addressed criticisms of grounded theory in the 3rd edition of this Handbook in some detail, so will only present those from within grounded theory here.
 Over the years, Glaser (2001, 2003) has changed his view and now states that researcher conceptualizes participants’ main concern.
 Structural-functionalism was the reigning theory of the 1950s. It invokes a biological metaphor, addresses the structure of social institutions, and evaluates how well they accomplished key societal tasks such as socializing children and controlling crime. Structural-functionalism assumes consensus between individuals and segments of society, studies social order, and emphasizes social roles within institutions (see Merton, 1957; Parsons, 1951).
 Pragmatism not only informed Strauss’s work, but also he stayed within and developed the pragmatist tradition through symbolic interactionism. This perspective emphasizes interaction, language and culture as shaping the construction of meaning and action. It assumes a dynamic relationship between agentic, reflective actors and society and thus sees social institutions and society as constructed, not as givens (see Blumer, 1969; Reynolds and Herman, Strauss, 1969, 1993).
 Fielding and Cisneros-Puebla (2010) integrate CAQDAS and GIS methods in an innovative mixed methods approach.
 These authors affirm that “qualitizing” quantitative data also occurs.
 Schwalbe et al’s (2000) article and Harris’s (2001; 2006a; 2006b) studies exemplify how inequalities are enacted.
 At the time I worked on this analysis, the literature contained many studies that used marginalization as a significant concept. However, authors left its meanings implicit and understood rather than taking them as problematic.
 This last point resonates with Patrick Biernacki’s (1986) grounded theory of natural recovery from heroin use. Based on his findings, Biernacki constructs an analysis of identity. Relinquishing heroin use without treatment turned on the significance of having and maintaining conventional identities.
 For a grounded theory of how school rules are enacted, see Thornburg, (2007).
 The reciprocal effects of interactional dynamics come into play here. Not all children who were ostracized, were bullied. Dixon found that temporary ostracism by peers kept some children’s angry outbursts in check and manageable.
 See Charmaz (2005) for these critiques and my responses.