An Exploration of Key Issues in the Debate Between Classic and Constructivist Grounded Theory

Anne O’ Connor, National University of Galway, Ireland
Barry Carpenter, University of Limerick, Ireland
Barry Coughlan, University of Limerick, Limerick, Ireland

Abstract

In this paper, the debates and discourse between Classic GT or Constructivist GT are explored. The aim of this paper is to evaluate the various claims in a critical manner by revisiting the original discourse outlining these approaches. The importance of maintaining a reflective, neutral stance while examining the arguments and evidence for the claims on both sides of this debate is emphasized.  In the final analysis, the view taken by these researchers was that valid arguments could be made to support Classic and Constructive GT approaches. The rationale for choosing a Classic GT methodology is outlined. Guidelines to support novice researchers in their task of choosing the most appropriate GT approach are suggested.

Keywords: classic grounded theory, constructivist grounded theory, literature review, participant voice, epistemological assumptions

GT was first described in 1967 by Glaser and Strauss and continues to be an evolving methodology with a number of iterations, ongoing debates, discussion and controversies with many researchers strongly identifying with one or other side in these debates (Glaser & Strauss, 1967). This paper will focus on the debates between Classic and Constructivist GT.

A challenge for novice researchers attempting to distinguish between approaches in GT is that the research designs share many core features and procedures (Bryant & Charmaz, 2007). While proponents of each approach emphasize their differences, the significance of these differences are often unclear to the GT novice. On first examination, it is difficult to identify the differences in the research process as the same terminology can be used to describe different processes or indeed the same process.  These issues are additionally confusing for a novice researcher as it is difficult to gain initial clarity on the importance of the various differences when comparing approaches.

A second challenge for the novice is the tendency for authors on either side of these debates to adopt polarized positions. There is a natural tendency in these debates for each side to emphasize their differences and the unique offerings of their approach while minimising the shared similarities. It can be difficult for a novice, who is unfamiliar with the area and in search of certainty, to retain a neutral stance and resist being swayed by the passion of the arguments rather than their rationale and content. The need for each researcher to appraise the arguments critically within the context of their own research aims is crucial.

A challenge for the novice is the way in which “competing” authors represent the opposing proponent’s writings.  Novice researchers must be careful not to accept these views uncritically. One example from this author’s experience was the example of Charmaz’s (2008) claim when referring to Glaser’s (1978) book Theoretical Sensitivity suggesting that “the abstract terms and dense writing Glaser employed rendered the book inaccessible to many readers” (p. 513). A reading of this book found that this claim was not supported as the terms used in the writing were clear and the writings was found to be logical, clear, and accessible.

While various authors used arguments from their proponents’ work to support their arguments, the researcher notes that quotes were occasionally taken out of context or did not reflect the full complexity of the original author’s thinking or certain researcher’s interpretation of the author’s position. Such observations emphasized the importance of critical appraisal of all sides of this debate and the importance of reading original writings rather than relying on the interpretations of others on them.

For a novice, who is seeking clarity, it is tempting to adopt one approach above the other quickly and accept all arguments in favour of that approach as his or her own. Managing the tension between the need to read and reflect deeply on the methodology and the time pressure inherent in the research process may result in researchers feeling under time pressure, and making uninformed decisions in favour of one approach over another. Fernandez (2012) cautioned that researchers who do not attain a wider understanding of the epistemological assumptions of their approach are more likely to be confused about their basic assumptions and develop a poor research design that may subject to internal inconsistencies.  Moreover, it is also argued that a deeper understanding of GT methodology enables a researcher to be more flexible in his or her use of the method while maintaining a consistent and cohesive research process.

In this paper, the debates and discourse between Classic GT and Constructivist GT will be explored. The authors in this paper aim to evaluate the various claims and counter claims in a critical manner and revisit the original discourse outlining these methods. The authors take a reflective, neutral stance while examining the arguments and evidence for the claims on both sides of this debate.  In the final analysis the view taken by these researchers was that valid arguments could be made to support Classic and Constructive GT approaches. The rationale for choosing a Classic GT is outlined with guidelines for others when faced with this choice.

Comparing Constructivist and Classic GT

Constructivist GT and Classic GT are underpinned by two distinct paradigms. While they share many similarities, there are some notable differences which are reflected in their epistemological stance with related divergence on their research process and product. In the following section, the authors delineate these differences and similarities and outlines critical issues

           Epistemology. Epistemology is the branch of philosophy that explores the origin, nature, and methods of knowing and the limits of human knowledge. It therefore follows that a researcher’s epistemological stance has a significant influence on his or her choice and use of research methodology.

In a conversation with other gounded theorists, Fernandez (2015), as cited in Walsh et al. (2015), emphasized the importance of clarity and sound understanding of epistemological issues as they pertain to each research project as confusion in this area can lead to a poor research design and subsequent problems. As a researcher’s epistemological stance can inform numerous aspects of a project including the role of the researcher in the study, data collection methods, and analytic techniques, he argued that achieving epistemological clarity will lead to “well-defined and epistemologically congruent research outcomes” (Walsh et al., 2015, p. 587).

Constructivist GT and Classic GT hold divergent views on the epistemological underpinnings of their approaches.  Carter and Little (2007) explained that while these different epistemologies “represent internally coherent and workable approaches to research practice” they are mutually incompatible (p. 1320). O’Connor, Netting, and Thomas (2008) contended that the Classic GT approach is based on positivist, objectivist assumptions while the constructivist approach “is based on interpretivist, subjectivist assumptions” (p.42).

Classic GT epistemology. Classic GT theorists argue that Classic GT is flexible in its epistemological assumptions and therefore suitable for researchers from a wide variety of epistemological approaches (Holton & Walsh, 2016). Glaser (1992) argued that the epistemological basis of Classic GT is neutral as its “methods work quite well for analysing data within the perspective of any discipline” (p. 18). In an article describing conversations between Classic GT experts, Holton described Classic GT as “epistemologically flexible” (Walsh et al., 2015, p. 586). She posited that by utilizing the full analytic process of Classic GT, a researcher can use all types of data and a variety of epistemological approaches. Urquhart (2013) further outlined how Classic GT can be used successfully by interpretive, positivist, and critical realist researchers.

However, many authors described Classic GT as having an objectivist stance which adheres to the existence of an objective reality that can be studied and understood, although imperfectly (Charmaz, 2014). Glaser (1992, 2012) tended to avoid the dialogue on epistemologies resisting the pressure to delimit Classic GT’s epistemological assumptions. In conversation with others, he suggested that “for all the lofty academic talk, you can take GT whichever way you choose [explaining that] GT is just a set of steps that take you from walking in the data knowing nothing to emerging with a conceptual theory of knowing how the core variable is constantly resolved” (Walsh et al., 2015, p. 594)

Constructive Grounded Theory epistemology. In describing the epistemological assumptions of Constructivist GT, Charmaz (2014) argued that all knowledge is constructed and that reality is fluid and subject to changes based on a participant’s construction of it. She argued that Classic GT is based on an “outdated assumptions of an objective external reality, a passive, neutral observer, or a detached narrow empiricism” (Charmaz, 2014, p. 13). She described the constructivist perspective on research findings as constructed rather than discovered. Constructivist GT is predicated on the assumption that the researcher is an active agent in that construction and, as such, his or her position, privileges and perspectives are acknowledged as impacting the construction of knowledge in the research process. Thus, Charmaz (2014) claimed that “relativism characterises the research endeavour rather than objective, unproblematic prescriptions and procedures. Viewing the research as constructed rather than discovered fosters researcher’s reflexivity about their actions and decisions” (p. 13). Birks and Mills (2015) suggested that constructivist grounded theorists adopt a relativist position “where reality is relative to a conceptual framework, paradigm, form of life and is so constantly reformulated as a fluid construction” (p. 51).

Comparing epistemologies. Walsh et al. (2015) suggested that in its first iteration, Classic GT provided the researcher with an approach that could be used with any data and by researchers holding different philosophical assumptions.  The authors argued that, for this reason, GT has been viewed and described from a variety of perspectives which has led to a variety of theoretical products including analytic, explanatory, and predictive theories, and concluded that the tendency to emphasise just one application of GT in the form of one approach has diminished an overall appreciation of the scope and reach of Classic GT.   Holton and Walsh (2016) suggested that recognizing the various “researcher backgrounds, trainings and philosophical assumptions” (p. 16) enables one to understand the way in which GT has been used by many different researchers around the world.

Glaser’s views on the debate on epistemology can be summarized in the following quote:

GT is just a stupid little method. That’s all it is. The epistemology is irrelevant. It’s how you use it. GT is based on a concept indicating method which has been used for years in psychology. You get concepts out of indicators and the interchangeability of indicators and you get a theory. That’s it. (Glaser & Tarozzi, 2007, p. 27)

Research Process. Whichever approach is used, there is broad agreement within GT researchers on the essential characteristics of a GT study. These agreed procedures include simultaneous data collection, coding and memo writing, the use of the constant comparative method, theoretical coding, theoretical sampling, theoretical saturation and the importance of theoretical sensitivity (Hood, 2007; O’Reilly, Paper, & Marx, 2012; Urquhart, 2013; Wiener, 2007).

However, several distinct differences between Classic GT and Constructivist GT exist on a range of other processes within extant literature including the role of the researcher, procedures for use of literature, research questions, interview techniques and coding procedures.

Role of researcher. Both approaches emphasize the importance of the researcher remaining open to patterns identified in the data and being mindful of the potential impact of their own preconceptions on the research product. However, these preconceptions are managed differently within Constructivist and Classic GT approaches.

Charmaz (2014) suggested that researchers should actively engage in strategies that reveal preconceptions by taking a reflexive stance. She argued that while a researcher may come to the initial coding with certain preconceptions that act as starting points for looking at the data, they can only be adopted as codes when the data supports the codes. In this way, the preconceptions or professional concerns of the researcher only become part of the analysis if they are also reflected in the data.

In various writings, Glaser (1978, 1998) argued that a researcher’s preconceptions are like any other variable in the research and are subjected to the same analytical process so that their impact is managed through this process (Glaser & Strauss, 1967). Therefore, the researcher does not need to engage in identifying his or her preconceptions to avoid influencing the data analysis since the argument is that if these preconceptions are not reflected in the data, they do not become part of the final theoretical product. In this way, he argued the research product or GT is based on the patterns discovered in the data.

Within both approaches, the preconceptions or professional interests of the researcher only emerge in the theory when they are reflected in the data. However, constructivist grounded theorists engage in a process to identify their preconceptions whereas classic grounded theorists do not.

Within Constructivist GT, the role of the researcher is formulated as an active one in a process of co-constructing the final research product with participants.  Charmaz (2014) argued that researchers are part of the world they study and thus construct their theories through their “past and present involvements and interactions with people, perspectives, and research practices” (p. 17).

On the other hand, Glaser (2012) argued that the elevation of the researcher’s role to that of co-constructer of the data has the potential of placing more value on the researcher’s interpretation than the participants, thereby creating a situation in which the data is forced.

In response to the criticism levelled by Charmaz that classic grounded theorists are distant and objective, Glaser (2012) argued that the aim of applying GT procedures is to be as objective as possible. Glaser presented an alternative perspective of objectivity which ensures that the final conceptualization or theory is objective. Rather than conceptualizing distance as the space between the researcher and participant, he spoke about distance in terms of the final conceptualization or abstraction moving away from the raw data to achieve a broad theoretical explanation. He explained that that the more distance one achieves between data and abstraction the more explanatory power the theory will have (Glaser, 2012).

In reflecting on the debate in this area, there are clear distinctions between both approaches on the role of the researcher during the research process; within the Constructivist GT approach, the researcher is viewed as an active co-constructor of the final theory; within the Classic GT approach, the aim is to minimize the impact of the researcher’s preconceptions on the final theory through the application of essential GT analytic processes.

While both approaches agree on the centrality of properly representing the participant’s voice, both claim that the other does not adequately ensure the participant’s voice is reflected in the research product. Glaser (2012) asserted that by elevating the researcher’s role in the construction of the theory, there is a danger of placing as much or more value on his or her interpretation than that of the participants. He argued that the focus on mutual understanding and joint interpretations in the constructivist approach leads to a need to create agreement between the participant and researcher on the emergent knowledge; in other words, a drive towards accuracy and description. He argued that in this quest for accuracy, the latent patterns underlying multiple, often seemingly disparate participant perspectives may be lost (Glaser, 2012).

Charmaz (2014) argued that by not identifying and reflecting on the researcher’s preconceptions, there is a danger that these preconceptions will have unaccounted influences on the analysis. For this reason, she argued that researchers need to reflect and identify their preconceptions as part of this process.  However, both agreed that the application of the essential procedures of Classic GT ensures that any preconceptions of the researcher will only appear in the theory if they are grounded in the data collected from participants.

Role of Literature. Evans (2013) summarized the differences between Classic GT and Constructivist GT on the role of literature as follows: as the starting point for a Classic GT is a desire to explore a substantive area with no predetermined research questions prior to the study; it does not begin with a literature review. In contrast, a Constructivist GT begins with a literature review as a means to establish what is known and has been studied in the substantive area.

One of the key concepts relevant to the debate on the timing of the literature review is the basic GT tenet of theoretical sensitivity. Theoretical sensitivity describes the researcher’s ability to recognize and extract the essential elements relevant to the emerging concepts, categories, and theory from the data.

Classic Grounded theorists argue that one of the key first steps in gaining theoretical sensitivity is embarking on the research with as few preconceptions or predetermined ideas as possible. In particular, Glaser and Strauss (1967) argued against the use of logically deduced a priori hypothesis. The rationale is that if a researcher has already read and reflected on the literature in the area, which includes existing theories, it will negatively affect his or her ability to maintain theoretical sensitivity since he or she will not be able to “record events and detect happenings without first having them filtered through and squared with pre-existing hypotheses and biases” (Glaser, 1978, p. 3).

Glaser (1978) suggested that while initially avoiding the literature on specific theories related to the areas of enquiry, researchers should explore the wider literature to become familiar with a variety of theoretical codes to support their developing knowledge on theory building. Theoretical sensitivity can be increased by the researcher familiarizing themselves with a wide range of theoretical perspectives and the construction of theories in general. As the analytic phase of theoretical coding reaches completion, a literature review can then be undertaken. In this way, the emerging theory is integrated into the existing literature and its generation has not been unduly influenced by existing theory and literature.

It is thus argued that delaying the literature review in this way reduces the possibility that the data will be forced to align with preconceived concepts and allows time for the core category to emerge which will ultimately generate a more focused and effective literature review.

Within Constructivist GT, the researcher is encouraged to become familiar with the literature prior to data collection. Charmaz (2014) made the point that it is unrealistic to expect that researchers will start their research without holding particular perspectives and knowledge about its focus. She argued that the stricture to delay the literature review implies that researchers are uncritical in their reading and are easily persuaded by it. Rather she argued that researcher’s ability to develop theoretical sensitivity is predicated on his or her familiarity with relevant literature.

She suggested that the argument of when to conduct a literature review tends to miss the crucial and most important point that the researcher should tailor the final version of their literature review to fit their particular project and its findings. This perspective is shared by Glaser (1978, 1998) and both scholars  emphasized the importance of properly integrated the emergent theory within the existing literature (Charmaz, 2014; Glaser, 1978, 1998).

While it is recommended that the timing of the literature review is delayed until the theory has been generated, there is an acknowledgment by Classic grounded theorists that there are practical reasons why a researcher must demonstrate some familiarity with the literature in the initial stages of their research, such as in supporting their research proposal (Glaser, 1998, Holton and Walsh, 2016). Nonetheless, they stressed that a literature review conducted after the identification of the core category is more efficient and relevant.

In conversation Fernandez (2015) also argued that the work of the researcher is to integrate the emergent substantive theory into the existing formal theory thus embedding the research findings in the existing literature. He asserted that many researchers do not adequately address this step in the research process (Walsh et al, 2015). Glaser (1998) argued that the debate and focus on the timing of the literature review reflects a basic misunderstanding of the inductive process. Within this approach, the role of literature and existing formal theory are supported but not until the data has been allowed to direct the research focus (Glaser,1998).  

Research questions. Within a Constructivist GT, the researcher develops a set of research questions prior to data collection. These questions are based on the literature review and direct the choice of data collection methods and development of interview guides. These questions may change during the research as the researcher discovers more significant or relevant questions.

Within a Classic GT approach, the researcher does not develop a prior set of research questions; rather the researcher seeks to approach the substantive area with a broader question that facilitates the participants to speak about their experiences (Glaser, 1998). As the research progresses through the stages of coding, the researcher may ask more direct questions relating to the already-generated categories.  It is argued that by approaching the substantive area in this manner, the researcher remains more open to the emerging patterns from that data and avoids the risk of steering participants down certain routes too early in the research process, thereby forcing the data.

Interview Techniques. While both approaches adopt similar guidelines on qualitative data collection methods, they adhere to two distinct strategies in preparing for interviews and in their stance on the question of devising an interview guide. Charmaz (2014) recommended that new researchers develop a detailed interview guide to enable them to gain clarity on the type of information they seek to address their research questions. The purpose of the guide is also to avoid the use of “awkward, poorly judged questions potentially based on unexamined preconceptions” (Charmaz, 2014, p. 63).

Glaser (2012) argued that the type of interview described by Charmaz is problematic: long guided interviews can force or lead the interviewee in certain directions and impose interview bias on data. Rather, he favours a passive, non-structured interviewing style in which constructivism is kept to a minimum.  Initially, the interviewer is merely receptive; it is only later in the process, when theoretical sampling dictates more focused interview questions, that the researcher’s input becomes more targeted. This more active interviewing is guided by analysis of data as opposed to the biases or experiences of the researcher.

As an illustration of the above point, one Constructivist GT study’s approach to the process of co-construction in data collection is presented; the authors described their interview process as follows:

Initial questions were broad and open-ended; as we interacted with the data and as categories were co-constructed, we adjusted the range of topics to gather more specific data to develop our theoretical framework. For example, each time a mother introduced a new topic or idea such as money, time, interpersonal conflict, or unique strategies, we added this insight to future interviews to see if it was common for other mothers. (Walsh, Meagher-Stewart, & Macdonald, 2015, p. 529)

This example, which the authors purported follows a Constructivist GT methodology, shows how subsequent interview guides are shaped based on earlier interviews. While this process may be presented as an example of theoretical sampling, it could equally be viewed an example of forcing the data. As each participant introduces new concepts, they are then repeated in the data by the researcher thereby introducing them in subsequent interviews.

These additions to the interview protocol are arguably not based on the analytic process of identifying gaps in the data (theoretical sampling) as much as the decision to introduce them based on their novelty.  If these concepts subsequently assume analytic relevance, it is difficult to determine whether their repetition in the data was due to the intervention of the researcher or whether the concepts would have emerged without the researcher’s influence. Thus, an inherent issue with co-construction during the data collection process would appear to be distinguishing between the voices of the participant and the researcher. However, it should also be noted that Charmaz continually emphasizes the importance of researcher’s awareness of forcing data and of deploying techniques which minimize this risk.

Glaser (2012) also cautioned that Charmaz’s description of the data collection phase as a potential therapeutic encounter between the interviewer and interviewee is confusing therapy with research.

On the other hand, Charmaz (2014) characterised the Classic GT approach to their participants as cold and distant. Glaser and Strauss (1967) earlier writings describing GT methods reflected the widely-held view, at the time, of the participant as solely a source of data. The language used to describe participants was typical of an era which had not developed our contemporary awareness and sensitivity to ethical considerations or, the rights of participants and responsibilities of researchers within the researcher participant relationship. 

Coding. Both approaches subscribe to a cycle of coding that moves from line by line analysis of raw data to theoretical coding and sorting of categories using analytic memos in support the development of theory. Within Constructivist GT, three distinct phases of coding are described: initial coding, focused coding, and theoretical coding (Charmaz, 2014). Classic GT has two main coding phases: substantive (including open and selective) and theoretical (Glaser, 1978, 1998; Glaser & Strauss, 1967).

Both authors (Charmaz and Glaser) described open and initial coding in similar terms: the focus is on breaking down or fragmenting the data to identify patterns and recurring concepts or categories. Selective and focused coding are also similar but within Classic GT, the researcher only moves to selective coding following the identification of a core category. Also, unlike Classic GT in which just one core category is developed, within Constructivist GT, several core categories can be developed and described in the final document.

When describing theoretical coding, Charmaz (2014) often referred to and supported Glaser’s descriptions of this process and seemed to describe a similar process. However, the emphasis in Constructivist GT is in merging and grouping concepts whereas in Classic GT it is on first identifying the relationship between the categories and the core category and then identifying the relationships between categories.  Within Constructivist GT, the process of theoretical coding would seem to be optional and perhaps at the researcher’s discretion while theoretical coding is an essential element of the Classic GT research process.

Research product. Charmaz (2014) pointed out that it can be difficult for those new to GT to clarify what “stands as grounded theory in any version” (p. 15). Within the Classic GT tradition, the GT is defined as “the systematic generating of theory from data, that itself is systematically obtained from social research” (Glaser, 1978, p. 2). The aim of Classic GT is to develop a multivariate comprehensive theoretical explanation, as Glaser put it.  GT simply involves “the generation of theories from data” (Walsh et al., 2015, p.593).

Within the constructivist tradition, the grounded theory is the result of a process of data collection and analysis as created by shared experiences and relationships between the participants and the researcher, and is a conceptual description of this process. Along with theorizing the interpretive work undertaken by participants, within this tradition, the final theory is the result of a co-construction of knowledge and is not separate from the researcher’s view and cannot exist outside of this view (Charmaz, 2014).

Rationale for Choice of Classic GT

In a contribution to a conversation on GT Fernandez (2015) cautioned that researcher must take responsibility for gaining a clear understanding of the epistemological assumptions underlying their chosen methodology (Walsh et al., 2015). The process of choosing the GT approach is a complex exercise which required extensive reading and reflection.

When faced with the choice of Classic or Constructive GT, given that arguments could be credibly made for both approaches, the choice is not based on a determination of which was the best approach but rather on which approach best suits both the researcher and his or her study. The individual researcher’s perspective’s influence is acknowledged by Heath and Cowley (2004) who contended that qualitative research using grounded theory is a

cognitive process and that each individual has a different cognitive style.  A person’s way of thinking, and explanation of analysis, may seem crystal clear to someone with a similar cognitive style and very confusing to another person whose approach is different. (p. 149)

For these researchers, in considering the similarities and differences between the two approaches, the Classic GT approach was ultimately chosen based on the following rationale:

Epistemological flexibility. While some authors suggested that Classic GT is restricted to one epistemological stance,  others contended that it is an approach which may be used by researchers from diverse epistemological perspectives. Fernandez (2012) emphasized it is more important that the researcher understands which perspective he or she is adopting and ensures that all research decisions are consistent with this perspective. In reviewing and reflecting on the epistemological assumptions, the authors have determined that the closest to their stance is that of critical realism as described by Oliver (2012) as an alternative to the objectivist or constructivist view in GT. Within critical realism, the existence of an objective reality exists independently of our thoughts and perceptions. It is further argued that the description of this reality is mediated and filtered through our use of language: our meaning-making and our social context. The use of a Classic GT methodology is consistent with this epistemological stance.

Participants’ voices. The second argument which supported the choice of a Classic GT concerned the participants’ perspectives and experiences that would be reflected adequately in the analysis and final product. Two issues were considered relevant to this concern:

Avoiding preconceptions. The potential influence of an early comprehensive literature review on the researcher’s preconceptions was of some concern. The dangers of this process unduly influenced the choice of research questions; its effect on the analysis were considered. While acknowledging the constructivist position that researchers should not be depicted as uncritically accepting their readings and being unaware of their potential effect in the research process, nonetheless the unconscious bias that existing theory could exert on their work is not, in the authors’ opinions, adequately addressed within the constructivist approach.    While there are suggested strategies to counteract researchers’ bias, it was the view of the authors that the arguments in favour of a delayed literature review were better supported; we acknowledged that Charmaz and Glaser emphasized that from their perspective, the most important point is that the final theory is well integrated within existing literature.

Role of researcher. The second issue was the role of the researcher in the research process. Given that the role of the researcher within Constructivist GT is co-constructor of the final research product, the authors considered the danger of undue influence on the analysis.  The argument is that the researcher’s perspectives will only find their way into the final theory if they emerge during the analysis. However, in the authors’ opinions, these arguments do not fully address the possibility of the data being unduly influenced prior to analysis in the collection phase.  Due to these misgivings, the authors accepted the argument that the role of the researcher within a Constructivist GT process could lead to undue influence on the data or forcing the data.

Analytic guidelines. When considering both approaches, the comprehensive, clear, and flexible guidelines offered by experts in Classic GT were an additional incentive to select this approach. In comparing the guidelines for both approaches, the authors concluded that the procedures and guidelines available in a variety of writings that outline the Classic GT approach would better support a researcher through all the stages of the research process from data collection through to the discovery of a multivariant integrated theory (Glaser, 1978, 1998; Glaser & Strauss, 1967).

Potential Framework for Choosing GT approach

In their process of choosing a GT approach, the authors accepted the caution expressed by Jacoby, Jaccard, and Acock (2011) that rather than seeking to divide and polarize we should strive to develop a diverse set of research tools to complete the research project.  They pointed out that positivist and constructivist perspectives are not necessarily conflicting; the reality is that things rarely fit into simple categories. Fernandez (2012) also called for researchers to concentrate on understanding the differences between GT approaches and “quit the private and public bickering – bury the territorial hatchet” (p. 27).

Based on the experience of the authors, scholars struggling to select a GT approach for their research would be best advised as follows:

• Read the original descriptions of the approaches under consideration as this literature will facilitate a critical appraisal of the arguments for and against each approach.
• Despite the pressure to do otherwise, try to retain a neutral stance while reflecting on the decision. Since all these approaches have merits, it is not merely a matter of which is better, but rather a decision based on individual philosophy of science and research interest.
• When considering the choices, adopt a logical approach to decision making, and compare each approach on the dimensions most relevant to you and your project
• When reading this literature first, expect to be confused and frustrated, as the tenor of the debate can at times seem very personal and the similarity of terminology between the approaches confusing.
• Do not rush the process. Allow plenty of time to read and reflect. Time spent on this process will not be wasted as clarity around research design will lead to a clear plan and internal consistency between the elements of the research.

Researchers in this area are continually refining their approaches and recommendations on practice. Therefore, a novice will not find easy answers in developing a research design and process. Rather, it is important to approach this method with a willingness to engage in a high degree of reflective practice prior to and throughout the overall research process.

Summary of Rationale for Choosing a Classic GT Methodology

The main areas of debate between approaches centre around the purpose of GT, the underlying assumptions of the researcher on the nature of the world and science, the position of the researcher in the study, the timing and role of the literature review, the development of research questions and interview techniques, and the coding and description of the research product.

Grounded theorists would argue that a researcher’s final decision should be based on his or her own epistemological stance and the needs of the specific research project. Based on an exploration and reflection on the Classic GT and Constructivist GT methodologies in terms of their fit for the authors’ epistemological perspectives and research project  the Classic GT research design was chosen. The choice was based on epistemological stance, concern that participant’s voice would be adequately reflected in the theory and pragmatic considerations concerning the quality of the directions available regarding the practical application of the research design.

In reflecting on epistemological perspectives, the authors concluded that as a critical realism stance most closely matched their perspectives the flexible approach of Classic GT would work best. The contention that the application of a Constructive GT may result in forcing the data by introducing researcher bias was explored and accepted. Finally, in a review of the “manuals” available to GT researchers, the authors concluded that those works pertaining to Classic GT were clearer and supported the authors through all the phases of the research. Of particular support were two primers developed by Glaser (1978, 1998), Theoretical Sensitivity and Doing Grounded Theory: Issues and Discussions, which provide clear guidelines for future Classic GT researchers.

References

Birks, M., & Mills, J. (2015). Grounded theory: A practical guide (2nd ed.). Los Angeles, CA: SAGE.

Bryant, A., & Charmaz, K. (2007). Grounded theory research: Methods and practices. In Bryant,A. Charmaz, K. (Eds)  Handbook of grounded theory (pp. 1-29). Los Angeles, CA: Sage.

Carter, S. M., & Little, M. (2007). Justifying knowledge, justifying method, taking action: Epistemologies, Methodologies, and methods in qualitative research. Qualitative Health Research, 17(10), 1316-1328. doi:10.1177/1049732307306927

Charmaz, K. (2008). Constructionism and the grounded theory method. In  J. A. Holstein, J.A., & Gubrium, J. F (Eds.), 2008 Handbook of constructionist research, (pp. 397-412). New York, NY: The Guilford Press.

Charmaz, K. (2014). Constructing grounded theory. Los Angeles, CA: Sage.

Evans, G. L. (2013). A novice researcher’s first walk through the maze of grounded theory: Rationalization for classical grounded theory. Grounded Theory Review, 12(1), 37 – 55.

Fernandez, C. (2012). Guest editorial, themed section. The Grounded Theory Review, 11(1), 7-28.

Glaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory. 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. (2012). Constructivist Grounded Theory? The Grounded Theory Review, 11(1), 28- 38.

Glaser, B. G. (2016) The grounded theory perspective: Its origins and growth. Mill Valley, CA: Sociology Press.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research: New York, NY: Aldine de Gruyter.

Glaser, B. G., & Tarozzi, M. (2007). Forty years after Discovery: Grounded theory worldwide. The Grounded Theory Review: An International Journal, Special Issue, 21-41.

Heath, H., & Cowley, S. (2004). Developing a grounded theory approach: a comparison of Glaser and Strauss. International Journal of Nursing Studies, 41(2), 141-150. doi:10.1016/S0020-7489(03)00113-5

Holton, J. A., & Walsh, I. (2016). Classic grounded theory: Applications with qualitative and quantitative data. Los Angeles, CA: SAGE Publications.

Hood, J. C. (2007). Orthodoxy vs. power: The defining traits of grounded theory. In Bryant, A. Charmaz,K. (Eds.). Handbook of grounded theory (pp.151-164). Los Angeles, CA: Sage.

Jacoby, J., Jaccard, J. & Acock, A. (2011). Theory construction and model-building skills: a practical guide for social scientists. New York, NY: Guilford

O’Connor, M. K., Netting, F. E., & Thomas, M. L. (2008). Grounded theory: Managing the challenge for those facing institutional review board oversight. Qualitative Inquiry, 14(1), 28-45. doi:10.1177/1077800407308907

O’Reilly, K., Paper, D., & Marx, S. (2012). Demystifying Grounded theory for business Research. Organizational Research Methods, 15(2), 247-262. doi:10.1177/1094428111434559

Oliver, C. (2012). Critical realist grounded theory: A new approach for social work research. British Journal of Social Work, 42(2), 371-387. doi:10.1093/bjsw/bcr064

Urquhart, C. (2013). Grounded theory for qualitative research: A practical guide: Los Angeles, CA: SAGE.

Walsh, A., Meagher-Stewart, D., & Macdonald, M. (2015). Persistent optimizing: How mothers make food choices for their preschool children. Qualitative Health Research, 25(4), 527 –539.

Walsh, I., Holton, J. A., Bailyn, L., Fernandez, W., Levina, N., & Glaser, B. (2015). What grounded theory is . . . A critically reflective conversation among scholars. Organizational Research Methods, 18(4), 581-599. doi:10.1177/1094428114565028

Wiener, C. (2007). Making teams work in conducting grounded theory. In Bryant, A. Charmaz, K. (Eds). Handbook of grounded theory (pp. 293-310). Los Angeles, CA: Sage.