An empirical appraisal of Canadian doctoral dissertations using grounded theory: Implications for social work research and teaching

JOURNAL OF TEACHING IN SOCIAL WORK https://doi.org/10.1080/08841233.2017.1386259 An Empirical Appraisal of Canadian Doctoral Dissertations Using Grounded Theory: Implications for Social Work Research and Teaching Morgan Braganzaa, Bree Akesson a , and David Rothwell b Faculty of Social Work, Wilfrid Laurier University, Kitchener, Ontario, Canada; bCollege of Public Health and Human Sciences, Oregon State University, Corvallis, Oregon, USA Downloaded by [99.252.168.12] at 11:50 09 November 2017 a ABSTRACT KEYWORDS Grounded theory is a popular methodological approach in social work research, especially by doctoral students conducting qualitative research. The approach, however, is not always used consistently or as originally designed, compromising the quality of the research. The aim of the current study is to assess the quality of recent Canadian social work doctoral dissertations implementing a grounded theory approach. Our analysis is based on the premise that presentations of grounded theory approaches in doctoral dissertations impact the conduct of teaching and future research and have direct implications for the legitimacy of qualitative research. Using Saini and Shlonsky’s Qualitative Research Quality Checklist, the authors appraised dissertations in terms of credibility, dependability, confirmability, transferability, authenticity, and relevance. The article concludes with implications regarding the quality of studies utilizing grounded theory approaches and consequences for future doctoral education and research. Canadian; doctoral education; teaching research First articulated by Glaser and Strauss (1967), grounded theory is one of the most widely used and thoroughly described research approaches in the social sciences, including social work (McKibbon & Gadd, 2004). A review of 248 Canadian social work doctoral dissertations found that of the 83% (n = 206) dissertations coded as qualitative or mixed methods, 22% used grounded theory (as compared to 26% qualitative description, 23% not classified, 12% phenomenology, 5% ethnography, and 2% case study (Rothwell, Lach, Blumenthal, & Akesson, 2015)). However, like other qualitative approaches, such as phenomenology (Norlyk & Harder, 2010), grounded theory is not always used consistently or as originally designed, potentially compromising research quality. The aim of this study is to appraise the quality of Canadian social work doctoral dissertations that use a grounded theory approach to advance debates on the quality of qualitative social work research. This article CONTACT Bree Akesson bakesson@wlu.ca Faculty of Social Work, Wilfrid Laurier University, 120 Duke Street West, Kitchener, Ontario N2H 3W8, Canada. © 2017 Taylor & Francis Group, LLC 2 M. BRAGANZA ET AL. Table 1. Elements of the Qualitative Research Quality Checklist. Qualitative framework Study setting Study design Sampling procedures Data collection Downloaded by [99.252.168.12] at 11:50 09 November 2017 Ethical issues Data analysis Findings Authenticity Fairness Promotion of justice ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Research purpose and question Appropriateness of qualitative approach Appropriateness and specificity of study setting Prolonged engagement Persistent observation Appropriateness of research design Sample selection Sample size and composition Description of methods Consistency of methods Range of methods (triangulation) Articulation of collection and analysis Audit trail Informed consent, privacy, confidentiality, and protection from harm Detail and depth of analysis description Matching quotes to concepts/themes Emergence of findings from experiences of sample Member checking Thick description Stakeholders’ involvement Stakeholders’ access to research process and benefits Enhancement of stakeholders’ reality Empowerment of stakeholders Note. Source: Saini (2011). provides a review of elements of grounded theory research among Canadian social work doctoral dissertations published between 2001 and 2011. To appraise the quality of the dissertations, we used (Saini, 2011) the Qualitative Research Quality Checklist (QRQC). The QRQC is a comprehensive assessment tool designed to assess the quality of any qualitative approach to research, such as ethnography and case study. It includes practical assessments (e.g., around the articulation of the research question, number of participants, ethical considerations), as well as epistemological considerations in the research design (see Table 1). Our analysis is based on the premise that presentations of empirically grounded theory approaches in doctoral dissertations have consequences for the teaching and conduct of research, as well as direct implications for the legitimacy of qualitative research itself (Norlyk & Harder, 2010). Our findings lead to a series of questions that have implications for teaching, as well as use of grounded theory in qualitative doctoral research. This process, however, raised more questions than answers for social work teaching and pedagogy, and we offer them as points of discussion for both methods courses and advising doctoral students. We wish to emphasize that the motive for this article was to attempt to assess the overall quality of grounded theory in social work doctoral dissertations, clarify the challenges in conducting grounded theory research, and provide steps to advance the practice of qualitative inquiry. In the Findings JOURNAL OF TEACHING IN SOCIAL WORK 3 and Discussion section, we use examples from actual dissertations to enable readers to follow how we came to our conclusions. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Assessing the quality of qualitative and grounded theory research Since the introduction of qualitative methods, there has been the need to verify qualitative studies through a careful assessment of methodological quality and rigor. According to Miles and Huberman (1994), “We have the unappealing double bind whereby qualitative studies can’t be verified because researchers don’t report on their methodology, and they don’t report on their methodology because there are no established cannons or conventions for doing so” (p. 244). In light of this, several scholars have made attempts to establish general guidelines for assessing the qualitative research process using notions such as rigor (Rubin, 2000; Tobin & Begley, 2004), trustworthiness (Krefting, 1991; Shenton, 2004), and authenticity (Erlandson, Harris, Skipper, & Allen, 1993; Lincoln & Guba, 1985; Whittemore, Chase, & Mandle, 2001). At the same time, others have argued against developing any standardized criteria for evaluating qualitative research. For example, Padgett (2008) felt that because qualitative research begins from the premise that knowledge is both created and contextual (as opposed to discovered), developing standardized criteria across different research contexts would be the antithesis to authentic qualitative inquiry. Along the same lines, Barbour (2001) noted, “If we succumb to the lure of ‘one size fits all’ solutions we risk being in a situation where the tail (checklist) is wagging the dog (the qualitative research)” (p. 1115). In other words, although quality assessment checklists may contribute to increased confidence in the quality of qualitative inquiry, they also can be counterproductive if rigidly followed. In light of these debates, there is no agreement on an actual tool to aid in assessing the quality of qualitative studies. In fact, there are more than 100 quality appraisal processes to assess the rigor and credibility of qualitative research studies (see, e.g., Creswell & Miller, 2000; Drisko, 1997; Forchuk & Roberts, 1993; Patton, 1999; Reid & Gough, 2000; Tracy, 2010). Although helpful, many of these tools “do not distinguish between different epistemological and ontological differences, theoretical frameworks, study designs, and standards for rigor, credibility, and relevance” (Saini & Shlonsky, 2012, p. 113). These tools, therefore, do not offer a quality appraisal process specific to grounded theory (and therefore attend to the inherent ideological tensions between the various epistemological approaches that influence the assessment of quality). Specifically, Glaser and Strauss (1967) introduced grounded theory as a response to the trend in positivistic approaches to social science research. Whereas much theory development was done a priori, before collecting and Downloaded by [99.252.168.12] at 11:50 09 November 2017 4 M. BRAGANZA ET AL. analyzing data, Glaser and Strauss argued for an alternative approach in developing theories. Indeed, as a methodology, grounded theory generated new theory from data, as opposed to testing existing theory (Strauss & Corbin, 1994). Nevertheless, the approach and rigor related to data collection, handling, and analysis created differences between Glaser and Strauss. Following an “epistemological divorce” from his colleague Glaser, Strauss forged an alliance with qualitative researcher Corbin to advance the grounded theory method (Strauss & Corbin, 1990). They were informed by pragmatism and symbolic interactionism. Later, Charmaz (2006) introduced a constructivist approach to grounded theory, arguing that there are multiple realities in the world and “generalisations are partial, conditional and situated in time and space” (p. 141). The epistemological approaches to grounded theory methodology alter the research design, influencing the assessment of quality, as well as necessitating the usage of a checklist that considers epistemology. Based on our appraisal, we hope this article may advance current debates on the quality of qualitative social work, as well as illuminate the inherent ideological tensions between positivist and constructivist positions. Method Sample Our sample was drawn from a database of 248 publically available social work PhD dissertations between 2001–2011 at 14 postsecondary institutions in Canada that offer a PhD program (Rothwell, Lach, & Blumenthal, 2013). Most dissertations were found in library searches using ProQuest. The doctoral dissertation is considered to be the signature pedagogy of doctoral education in social work (Anastas, 2012), representing the culmination of doctoral studies through the conduct and reporting of original empirical research that generates new knowledge to advance the discipline (Maynard, Vaughn, Sarteschi, & Berglund, 2014b). In this way, doctoral dissertations are an important barometer for understanding the current methodological trends in a given field. In comparison to other mediums, such as journal articles, doctoral dissertations were chosen due to their comprehensive nature. Although not exhaustive, we feel a 10-year period provides a reasonable sampling to gauge major features of contemporary social work dissertations (Maynard, Vaughn, & Sarteschi, 2014a). (For details regarding the original sampling frame from which our sample was drawn, please see Rothwell et al., 2015.) Of the 248 dissertations from the larger study, 46 dissertations were categorized as implementing a grounded theory methodology. Upon close examination of the sample, one dissertation was deleted from the sample due to being miscategorised as grounded theory, and six dissertations were eliminated because they were written in French (in which neither of the JOURNAL OF TEACHING IN SOCIAL WORK 5 researchers conducting the analysis was fluent). Hence, the final sample was 39 dissertations. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Appraisal tool To appraise the quality of the social work dissertations, we used Saini’s (2011) QRQC, a 25-point rating form. The QRQC was pilot tested for interrater agreement by master’s-level social work students, and modifications subsequently were made based on interrater agreement (Saini & Shlonsky, 2012). The QRQC evaluates qualitative studies in terms of their epistemological and theoretical frameworks, study setting, study design, sampling procedures, data collection, ethical issues, data analysis, findings, and authenticity. In addition, fairness and promotion of justice are included to evaluate studies for which the central purpose is to empower participants through participatory research (Saini & Shlonsky, 2012). Table 1 identifies these elements. (A more detailed overview of how these elements were operationalized can be found in Braganza, Akesson, & Rothwell, 2015.) Procedure We (the first two coauthors) conducted the data analysis independently. We reviewed the abstract, introduction, methods, and findings sections of all 39 dissertations, appraising the various elements of them according to the QRQC. When considering elements of grounded theory, we maintained an open stance, acknowledging the range between grounded theory approaches from the more positivist Glaser and Strauss (1967) approach to the more constructivist Charmaz (2006) model. We tracked our ratings independently, meeting every 2 to 3 weeks over a 6-month period to check interrater reliability and to develop consensus on any responses that differed, using a consensus-building process (Jones & Hunter, 1995). Each item was assigned two ratings—“applicable” and “addressed”—each of which consisted of three possible elements (yes, no, or unclear (Saini & Shlonsky, 2012, p. 170)). (Due to differences in epistemological and ontological stance, not all elements of the QRQC are “applicable” to qualitative studies.) The QRQC includes three columns for each element. The first column indicates whether the element is relevant to the particular study. Our determination of applicability was made based on (a) what the respondents stated they did and (b) their epistemological paradigm. In this study, the dissertation author’s epistemological lens, whether stated or implicit, was considered in order to assess the overall quality of the study and became a standard against which to measure. The appropriateness of the research design also was considered vis-à-vis the author’s epistemological paradigm. The second column indicates whether there is sufficient information in the 6 M. BRAGANZA ET AL. study for the reviewer to be able to assess whether the specific element has been “addressed.” A third column provides space for the reviewer to comment generally upon the study and the qualitative element being addressed. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Ethical considerations According to Padgett (2008), the use of labels “often obscures the blurriness of categories and can easily verge into name calling” (p. 89). Saini and Shlonsky (2012) therefore recommend that a qualitative methods expert be included on the research team. However, although we all have to varying degrees experience conducting quantitative, qualitative, and mixed-methods research, none of us claims to be an expert on qualitative research or grounded theory. Consequently, we made every effort to consult grounded theory researchers (and their scholarship) when we had questions about an element of this checklist. Findings and discussion The following section is organized by highlighting the most relevant findings related to how the surveyed dissertations addressed the following: (a) research question and design, (b) literature review, (c) methods used, and (d) the analysis and presentation of findings. (For a comprehensive presentation of the findings, please see Braganza et al., 2015.) A discussion of the findings also has been integrated in this section because these findings unearthed a variety of important questions and considerations. Hence, we believe these are best presented with the findings themselves to facilitate understanding. Research question and design Epistemological decisions Many dissertation authors clearly articulated their epistemological stance and related methodological decisions. Their doing so was helpful for assessing the quality of other criteria, such as research design and research question. In fact, it is recommended by Saini and Shlonsky (2012) that, when applying grounded theory, the epistemological lens of the research be considered against the research question and associated design. The quality of constructivist grounded theory cannot be assessed in the same way as classical grounded theory, because there are slightly different methodological decisions to be made. This finding also has larger implications for social work research and teaching. Epistemology and methodology intimately are connected in grounded theory. Perhaps social work instructors might teach epistemology and research methods together to highlight the connections. Doing so may JOURNAL OF TEACHING IN SOCIAL WORK 7 assist students in clearly articulating their epistemological and methodological decisions, as well as to avoid inconsistency between epistemological stance and the research design. With this said, not all researchers (or social work instructors) agree on the importance of epistemology in the first place. Is a clear (or overt) articulation of one’s epistemological stance even important in research? How problematic is a mismatch between epistemological stance and research design? Downloaded by [99.252.168.12] at 11:50 09 November 2017 Research question Our analysis pointed to variation in how the research question was presented in each dissertation, which of course is not necessarily a problem unique to grounded theory. To be specific, in many dissertations the question was stated differently throughout the dissertation (e.g., in the introduction vs. the methods section) or was not clearly stated in the first place. In Dergal’s (2011) dissertation, for instance, the stated research purpose was to “provide empirical evidence about who private companions are, what they do, and why they are” (p. ii), whereas in the introduction, the stated purpose was “to understand the reasons private companions are hired in long-term care facilities, and the role they play within the care environment” (p. 2). In some instances, there were several inquiries embedded in the research question. For doctoral students, who are often at the beginning of their research careers, it may be difficult to articulate one consistent research question, because their research may gradually be evolving during their doctoral studies. Instructors, therefore, might wish to consider whether variation is due to the evolution of students’ studies or difficulty in articulating research questions. Identifying the source of the challenge facilitates its resolution. We also examined the alignment between design and research questions. Generally, most research designs seemed to fit with the research question. In most instances, based on the purpose of the studies, an exploratory or qualitative design was indeed most appropriate. However, in at least one study, the author did not clearly justify why a qualitative approach was best. The few that did not seem to fit included mixed-methods designs that could have been “generally qualitative” rather than specifically grounded theory, and one that seemed to have methods that better reflected a case study methodology. These findings suggest that instructors need to continue their approach to educating students on how to select the most suitable research design for their research question. Reviewing the literature The inclusion of an extensive literature review in a doctoral dissertation is common and usually expected (Boote & Beile, 2005; Bruce, 1994), and thus all 8 M. BRAGANZA ET AL. Downloaded by [99.252.168.12] at 11:50 09 November 2017 of the dissertations reviewed included an extensive literature review. Nevertheless, the use of the literature challenged the assessment of quality of some of the dissertations when considering the research design and the usage of classic grounded theory methodology (Bryant & Charmaz, 2007), with its origins in the work of Glaser and Strauss (1967). Classic grounded theory was developed at a time when social science researchers did not have to satisfy human subjects ethics boards (Oktay, 2012). Therefore, when implementing a grounded theory approach, researchers were not required to consult literature prior to data collection and analysis. In this way, theory was authentically derived directly from the data while being uninfluenced by prior biases and ideas. These findings highlight the complexity of rigidly complying with both a methodology and doctoral requirements (Dunne, 2011; Nagel, Burns, Tilley, & Aubin, 2015). Indeed, one of dissertations discussed this reality, stating, Classic grounded theory methodology requires the researcher to approach the field without preconceived notions of the phenomena under study. . . . One enters the field interested in a general subject area and elicits from participants their main concerns or social problem. The nature of a doctoral dissertation is contrary to this classic notion of grounded theory in that a clearly set out question and full literature review are typically required, prior to the initiation of the study. (McLaughlin, 2006, p. 99) Social work instructors may need to consider whether (and how) it would be possible for doctoral students to fully comply with methodologies, as designed, and whether methodological purity (as opposed to methodological pluralism) is even the ideal to strive for. Furthermore, who has the power to decide whether a research project should have methodological purity—students or instructors? Such questions and considerations may be worth discussing in research methods courses. Methods used We found that dissertations used different terminology when describing the same elements (e.g., thick description, reflexivity, etc.). Further, not all were clear on how these things were actually done. These factors are a challenge within doctoral education, consistent with the findings of Nagel and colleagues (2015). For example, Lawson (2009) used the term “journals” to describe what we interpreted to be memos. But again, as readers we were unsure about the author’s use of this element of the methodology. This reality calls into question why authors are using these terms, whether they actually know what they mean, and how they are defining them. In other words, how these elements are identified and operationalized may be more important in qualitative research than we think. Instructors may consider acknowledging the challenges inconsistency creates for assessing quality and sharing JOURNAL OF TEACHING IN SOCIAL WORK 9 differing definitions of terminology in order to enter into debate about what these terms actually refer to. Moreover, qualitative research permits flexibility in terms of usage (or not) of elements like reflexive journals or thick description. Therefore, instructors might wish to guide their students in operationalizing and describing their decisions. Doing so will not only assist readers in assessing quality but also provide both budding and seasoned qualitative researchers direction on how to implement these elements. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Sample size and theoretical saturation Many dissertations in our appraisal were missing details on how the sample was created. Further, such studies often lacked detail on sample size and composition, or the justification for them. These omissions made it unclear who was involved in the research and why they were involved. Although grounded theory provides clear guidance about what constitutes theoretical and data saturation, how to apply them, and when to use them (Guest, Bruce, & Johnson, 2006), a number of dissertations applied theoretical and data saturation incorrectly. For instance, in some, it was unclear whether participants were sampled until data or theoretical saturation was achieved or if saturation was used as the justification for small sample size (see Brake, 2010; Dergal, 2011; Lach, 2004). In other words, a number of authors simply equated data or theoretical saturation with how many people agreed to participate. In one dissertation, a number of participants were invited to participate in the research, and all did, which raises questions about reaching saturation or achieving a target number of participants (Walker, 2005). In another, the final sample size was justified based on cost and time considerations. Later, in the same dissertation, the author claimed to reach theoretical saturation (McLaughlin, 2006). These findings suggest that doctoral social work faculty may wish to dedicate more time to conversations about sampling. Relatedly, instructors may consider linking sampling strategies to specific methodological designs so that students can understand that different methodological designs, such as grounded theory, have distinct sampling procedures. For example, theoretical sampling is a feature of grounded theory and is explicitly operationalized based on its connection to this methodology. Similarly, it was not always clear whether theoretical sampling—the process by which codes and analysis inform the researcher’s decision about where to next collect data (Glaser, 1978)—in fact was used as a rationale for sample sizes, or whether theoretical sampling was an appropriate sampling strategy to use to achieve an appropriate sample size and composition (Lawson, 2009). Such are important considerations for grounded theory in particular because some scholars have argued that sample sizes are expected to be larger for grounded Downloaded by [99.252.168.12] at 11:50 09 November 2017 10 M. BRAGANZA ET AL. theory research designs than for other qualitative designs, such as case study, ethnography, and phenomenology, to account for data saturation, theoretical saturation, and theoretical sampling (Saini & Shlonsky, 2012). This raises considerations for social work instructors such as whether it is possible for students to successfully achieve certain sampling features, like a large sample size or theoretical sampling, when they are faced with time and resource constrictions. Elements of sample size and saturation must also consider factors such as hard-to-reach populations (and other designs relative to the research project), which ultimately may require or result in a smaller sample size. Pursuing hard-to-reach populations is common in social work, and in such situations a smaller sample size does not mean that data saturation, theoretical saturation, and theoretical sampling cannot (or will not) be achieved, or that sampling was somehow flawed because the sample size ought to have been larger. Nevertheless, this finding raises questions about whether adopting saturation as a generic marker of quality is appropriate given that its meaning has evolved and transformed in a way that has been unchallenged in the literature (O’Reilly & Parker, 2012). How, then, can doctoral social work instructors explain the features of qualitative research that convey quality? One may also raise questions about whether these terms are actually unclear in the qualitative literature, or whether doctoral students are unclear about how to apply them. Findings like these suggest that perhaps social work research faculty should consider dedicating more time to explaining terms and to providing examples of how such terms are often applied differently in social work research. Triangulation We found a range of methods and strategies used to achieve different types of triangulation. In some situations, mixing methods was considered triangulation (Lawson, 2009). In other cases, comparing data from one interview against the data in another interview was deemed triangulation (Baylis, 2009; Munoz, 2011). In still another instance, “theoretical triangulation” was cited and named (Gowthorpe, 2010, p. 5). Many dissertation authors claimed triangulation was achieved, but it was difficult to tell whether this happened, and if so, whether it was relevant, based on the inconsistencies in definition and usage. The diverse strategies used to achieve triangulation, and different definitions of the term, also make it difficult to assess when and how triangulation occurred. Social work instructors might consider intentionally engaging with students in debates about what the definition, and therefore the usage and achievement of triangulation, ought to be. In these dissertations, was triangulation being used to understand the topic in a more nuanced way or to achieve some form of validity? In this way, instructors might encourage students to consider the purpose of triangulation in addition to its operationalization. JOURNAL OF TEACHING IN SOCIAL WORK 11 Transparency Many researchers do not explicitly say who collected the data, the dates when the research took place, or even who analyzed the data. Grounded theory, not unlike other research methodologies, is based on transparency and documenting every step of the process. How important is articulation, and how much detail ought to be provided in order to assess quality? These are important questions for the students to address. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Ethics Most dissertations spoke briefly about ethical procedures, as well as actual or potential ethical issues. Some provided more detail than others. Where dissertations focused on vulnerable populations or addressed very sensitive issues, we found that more information about ethical considerations clearly should have been included. In many dissertations, there was only a brief cursory overview of ethical procedures, such as providing informed consent, even when working with highly vulnerable populations, like adolescents with learning disabilities (Muskat, 2008). Faculty, therefore, might also consider whether enough time is spent discussing ethical issues in the classroom to ensure the centrality of this topic is understood. Analysis and presentation of findings Findings Per the QRQC, we aimed to determine if the findings emerged from the experiences of the sample. In one dissertation, it was difficult to say with surety the findings emerged from the experiences of the sample (Levy, 2004), because the findings section included fewer quotes than one might expect to find in a qualitative dissertation in order to demonstrate the connection between the subjective experience of the sample and the interpretation of the author. This example highlighted the importance of including direct quotations within a dissertation to support analysis and interpretation (Padgett, 2008). Dissertations with more quotes to illustrate themes, in addition to a detailed discussion about how data were derived from in vivo coding (Worthington, 2001), were easier to assess for quality. Thick description According to Ryle (1949) and Geertz (1976), thick description includes not only facts about the phenomenon under study but also the details, commentary, and interpretation of conceptual structures and meanings. (Thick description is more commonly used in ethnography than in grounded theory.) As one Downloaded by [99.252.168.12] at 11:50 09 November 2017 12 M. BRAGANZA ET AL. dissertation author states, “A thick description of the individual participants is contrary to grounded theory (Glaser, 1978) as analysts are interested in sampling and comparing emerging concepts, not individuals” (McLaughlin, 2006, pp. 94–95). However, thick description was assessed due to its inclusion in the QRQC and because, in a number of dissertations, either it was named as relevant or the author actually offered a thick description of the sample in the dissertation (Muskat, 2008; Timoshinka, 2008). Nevertheless, in many dissertations, insufficient details were provided to make the description of the sample “thick” (Monahan, 2001; Yan, 2002). In other words, the description of the sample was limited to biographical or demographic details. In others, the terminology used made it unclear whether thick description was relevant. For instance, some dissertation authors talked about offering “thick stories” of the sample (Lawson, 2009, p. 70; Munoz, 2011, p. 49; Zuk, 2009, p. 73), whereas others claimed to offer a “thick description of the situation” (De Zeeuw Wright, 2010, p. 64) or “thickness of the data” (Riedel Bowers, 2001, p. 110). Based on these definitions, it is unclear if a thick description of the sample even was relevant in many dissertations, although in some instances it was provided. This observation raises further questions regarding how thick description should be operationalized, and whether such operationalization should be consistent or flexible, and indeed whether thick description should be a feature in grounded theory studies. The findings suggest there are no straightforward or consistent answers for these questions. Member checking Member checking entails providing the research participants with a portion of the analyzed data in order to check its authenticity (Yanow & SchwartzShea, 2006). Most dissertations used member checking during the data analysis process, even though it is considered controversial within grounded theory (Charmaz, 2006; Glaser, 2002). Yet, we found that most examples of member checking did not provide a lot of detail (e.g., how many people were involved, how they were recruited), complicating our assessment. For instance, in two dissertations, justification was missing for why only two or three (among a larger number of participants) were selected to engage in the member checking process (Dergal, 2011; Gowthorpe, 2010). In one dissertation, it is not clear how many participated in member checking, even though the author talks about engaging in the process (Jenney, 2011). In another, the research justification for only selecting “the most articulate” participants to participate in member checking was not clear (Doiron, 2007, p. 29). In the remaining three dissertations, the authors described the theoretical rationale for engaging in member checking but did not provide enough detail on whether (McNeill, 2001) or how they engaged in the process (Helpard, 2010; Monahan, 2001), which raises questions about the rationale, process, JOURNAL OF TEACHING IN SOCIAL WORK 13 and articulation of member checking. For example, should member checking be used even if it is not traditionally included within a particular methodology, such as grounded theory? What should the member checking process look like within a grounded study? Downloaded by [99.252.168.12] at 11:50 09 November 2017 Identified biases In about half of the dissertations, the potential and actual biases of the researcher were identified and described, and there was a discussion about how these biases may have influenced the research design and/or process (Barnoff, 2002; Muskat, 2008). Engaging in reflexive practice throughout the research process is important in grounded theory because the emphasis of the constructivist paradigm is the construction of meaning (Charmaz, 2006). However, in several of these dissertations, researchers discuss their social location rather than their biases and do not identify the impact their biases could (or did) have on their research (Dill, 2010). For instance, in one dissertation the author identifies the various “selves” and experiences brought to the study but does not go on to describe how these may well have influenced the research design or process: “As the primary researcher in this study, therefore, it is appropriate to identify the intersecting selves that I bring to the study” (Milliken, 2008, p. 21). Our analysis raises questions for social work instructors to be prepared to field about reflexivity, to wit, Is social location always sufficient to identify biases? And is the integration of a reflexive journal—a written document detailing the researcher’s thoughts, feelings, and decisions—critical? Further, is acknowledging biases important when coming from a positivist paradigm? Developing theory The goal of grounded theory is to develop new theory; a grounded theoretical model or typology; or a deeper, more contextualized understanding of a phenomenon (Glaser & Strauss, 1967). Yet many of the dissertation authors did not explicitly develop any of the preceding from their data and analysis. It would seem here that grounded theory is being used more for exploratory purposes than for theory-generation purposes. This challenges the assessment of quality and also raises questions about why theory is not being generated. Perhaps developing a new theory is perceived as too ambitious for a doctoral dissertation project? One dissertation author stated that “sampling among divergent groups” is required in order to achieve a “full theory” (McLaughlin, 2006, p. 100), which, considering time and financial constraints, may not always be possible within the context of doctoral research. Some dissertation writers, however, did successfully collect data from “divergent groups” without producing new Downloaded by [99.252.168.12] at 11:50 09 November 2017 14 M. BRAGANZA ET AL. theory. It is possible that the methodology is attractive for exploratory or analysis purposes only, rather than for theory generation, particularly given the detailed steps that can be applied to either process. Indeed, a number of dissertation authors credit their usage of grounded theory to its “rigor” rather than its theory-building goal. Nonetheless, it is important for doctoral faculty to consider why theory is not being developed in studies utilizing grounded theory and whether this might relate to social work doctoral studies in particular. At a minimum, must a grounded theory study contribute to the theoretical understanding of a problem? The answer seems to be “yes,” otherwise the study may veer too far from the original goal of grounded theory as proposed by Glaser and Strauss (1967). The development of theory within this sample of dissertations raises further questions such as, Can the grounded theory methodology be used only for exploratory or analysis purposes, and if so, under what conditions? Indeed, is this usage considered a modification to the methodology or an inappropriate designation? Further, consideration should be given to those instances where authors claim to be using the grounded theory methodology, but only parts of the study (such as the analysis) make use of the design. How much can a methodology be altered before it is no longer that methodology? These questions have implications for instructors assisting students in choosing and utilizing selected designs to answer their research questions. Limitations First, our analysis was limited to the dissertation abstract, introduction, methodology, and findings sections. Although these portions compose the majority of the dissertation, and we often found all the elements within these sections, there was the possibility that elements may have been included elsewhere. For instance, the development of a theory, model, or typology may have been presented in the Discussion section, which was not assessed. Nevertheless, this limitation raises questions about what should be included in certain sections in qualitative research reports. Second, we reviewed only dissertations that were available in the data set, which included only dissertations that were publically available through the library. It is possible that other dissertations were produced that were not included in the data set. Third, the dissertation authors were not blinded during the appraisal process, which may have led to some element of observer bias. We attempted to control for this by including the reviews of two researchers, discussing any areas of potential bias, using the checklist to objectively rate quality, and using a consensus-building process. Fourth, although our sample of dissertations had many methodological strengths, this article focused primarily on what was lacking in order to underscore areas for future improvement and to comply with publishing space constraints. (A more detailed discussion of the JOURNAL OF TEACHING IN SOCIAL WORK 15 strengths of the dissertations is available in Braganza et al., 2015). Fifth, our analysis was solely of Canadian dissertations that were written in English. Finally, the implications for research and teaching raised in this article were derived from an analysis informed by the QRQC. It is possible that using a different assessment tool (or other analysis strategies) would have provided additional insights and implications. Downloaded by [99.252.168.12] at 11:50 09 November 2017 Recommendations for research and teaching Assessing the quality of qualitative research using a grounded theory design produced more questions than answers in relation to research and teaching. For instance, questions were raised about assessing the quality of qualitative research itself. As the author of one of the dissertations in our sample asked, What are the criteria for evaluating a methodology? Who decides on them? For what purposes? Through what processes do some and not others rise to prominence? These questions problematize the question of evaluation and reveal the processes of social construction beneath. And they are critical to ask, given that research results often become legitimized knowledge. (Brown, 2008, p. 92) Each dissertation might employ a different epistemological paradigm. This leads to the question of how to measure “quality” in qualitative research. It also offers a query about how social work faculty go about attending to and teaching on issues of quality. A dissertation that doesn’t meet the criteria of a checklist such as the QRQC might simply be reflecting a mismatch between the paradigm of the dissertation and of the approach to its appraisal. Furthermore, we acknowledge that a dissertation not achieving each and every one of the preceding standards should not suggest that the dissertation is not “good” or is of lesser quality. Nevertheless, difficult questions about the assessment of quality remain. Doctoral dissertations represent “a culmination and reflection of students’ personal and professional experience, substantive and methodological training, and professional socialization received in their doctoral programmes” (Maynard et al., 2014a, p. 269; see also Anastas, 2012). All doctoral social work programs, of course, provide research methods courses. Yet in qualitative method courses, grounded theory may be only a small part of a broad education on qualitative methods. To avoid some of the pitfalls uncovered through this research—such as having inconsistency between the epistemological stance and methodological design, or a misalignment between the research question and the research design—we suggest that social work programs consider offering full courses on particular methodologies such as grounded theory. Relatedly, we recommend considering whether to offer these courses earlier in a student’s program of study, such as before a literature review is completed, so students will be able to comply with the tenets of classic grounded theory. Downloaded by [99.252.168.12] at 11:50 09 November 2017 16 M. BRAGANZA ET AL. In addition, the finding that most dissertations did not generate new theory suggests that grounded theory may often be taught more as a data collection and analysis strategy than as a theory-generating process. The importance of developing theory may need to be more strongly emphasized in discussions of qualitative research methods within doctoral social work. On the other hand, perhaps some doctoral social work dissertations should aim to explore a social issue rather than generate theory? Many of the dissertation authors claimed to arrive at the research from a constructivist paradigm and therefore utilized Charmaz’s (2006) version of grounded theory. This illuminates several important considerations about the adherence to specific epistemological paradigms, and relatedly the student selection of research designs (e.g., case study, ethnography, grounded theory, narrative, or phenomenology). The first consideration this observation raises is whether the choice of epistemological paradigm and research design was influenced by the dissertation author’s advisor, discipline, and/or university, or whether it actually was voluntarily selected by the doctoral student. As Maynard et al. (2014a) note, “Given that dissertations are conducted under the advisement of a dissertation committee, the dissertation can also be viewed as a reflection of the interests and expertise of social work faculty” (p. 269). Hence, as Nagel and colleagues (2015) suggested, it is possible that the selection of research design is driven by social work faculty, especially when Canadian dissertations are compared to social work dissertations in other countries. For instance, Canadian doctoral students are more likely (65.3%) to use qualitative methods than students in Sweden (50%; Dellgran & Hojer, 2001), the United Kingdom (40%; Scourfield & Maxwell, 2010), and the United States (22%; Maynard et al., 2014). Having said this, in Gringeri and colleagues’ (2013) research of American qualitative dissertations that were completed between 2008 and 2010, it was found that grounded theory was a popular research design just as it is here in Canada (Rothwell et al., 2015). We would, in sum, recommend that social work instructors consider how these choices are presented and who holds the power to make the methodological choices. That many more authors used grounded theory from a constructivist rather than a more positivist, postpositivist, or pragmatic paradigm raises additional considerations with respect to epistemology. For instance, the trend raises questions about the popularity (and appropriateness) of particular epistemological paradigms in social work. Perhaps the positivist paradigmatic approaches used by Glaser and Strauss (1967) or Strauss and Corbin (1990) are less encouraged than the constructivist approach of Charmaz (2006)? Or perhaps positivism has less value in Canadian social work research than the constructivist approach? Or, perchance, the problems researched by these scholars simply lent themselves more appropriately to a constructivist use of grounded theory? Perhaps, as dissertation authors Dergal (2011) and De Zeeuw Wright (2010) claimed, grounded theory is simply a methodology Downloaded by [99.252.168.12] at 11:50 09 November 2017 JOURNAL OF TEACHING IN SOCIAL WORK 17 that ought to be considered for its rigor and its selection has little to do with epistemology. This suggestion, of course, raises questions about how authors approaching their research from various epistemological paradigms define “rigor.” Ultimately it is difficult to ascertain why approaching grounded theory through a constructivist paradigm would be more common in these dissertations. Yet, for the purposes of our discussion, this emphasis on a constructivist approach does raise larger epistemological considerations. For example, how should students be taught about the connection between methodological designs and each design’s epistemological stance? Should positions like positivism and constructivism be presented as equally valid positions by social work faculty? Or should instructors put forth positions that resonate more with their own epistemological stances? No matter the answer, we recommend that doctoral faculty engage students in dialogue about the variety of epistemological options and debates that inform the profession. Conclusions This study highlights the importance of assessing the quality of social work dissertations using a grounded theory approach specifically, and qualitative research more generally. It also points to the utility of a checklist, such as the QRQC, being applied to other research outputs such as articles in peerreviewed journals. More important, we believe our assessment of quality unearthed important yet challenging methodological and epistemological questions, which have significant implications for social work research and education. Although this inquiry perhaps raised more questions than answers, we believe these questions can stimulate fruitful class and mentor discussions. Acknowledgments We thank Lucy Lach for help with conceptualizing the study, as well as Jennifer Root and Victoria Burns for their helpful feedback and suggestions on an earlier version of this article. We also acknowledge the extremely helpful comments of the anonymous peer reviewers who made suggestions to deepen the final analysis. ORCID Bree Akesson http://orcid.org/0000-0001-6760-7198 David Rothwell http://orcid.org/0000-0003-0633-4071 References Anastas, J. W. (2012). Doctoral education in social work. New York, NY: Oxford University Press. Downloaded by [99.252.168.12] at 11:50 09 November 2017 18 M. BRAGANZA ET AL. Barbour, R. (2001). Checklists for improving rigour in qualitative research: A case of the tail wagging the dog? British Medical Journal, 322, 1115–1117. doi:10.1136/bmj.322.7294.1115 Barnoff, L. (2002). New directions for anti-oppressive practice in feminist social service agencies (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Baylis, P. (2009). The therapeutic alliance as experienced by children and its effect on neurological functioning (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Boote, D. N., & Beile, P. (2005). On the centrality of the dissertation literature review in research preparation. Educational Researcher, 34(6), 3–15. doi:10.3102/0013189X034006003 Braganza, M., Akesson, B., & Rothwell, D. (2015, November 5). Grounded theory in Canadian social work dissertations: Preliminary findings. Social Science Research Network. Retrieved from https://ssrn.com/abstract=2685945 Brake, E. S. (2010). A grounded theory study: Valuing aging - A model describing gerontological social work career choice (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Brown, M. (2008). She hits like a girl: Contexts and constructs of femininity, use of violence and living in group homes (Unpublished doctoral dissertation). Memorial University, St. John’s, Newfoundland, Canada. Bruce, C. S. (1994). Research students’ early experiences of the dissertation literature review. Studies in Higher Education, 19, 217–229. doi:10.1080/03075079412331382057 Bryant, A., & Charmaz, K. (Eds.). (2007). The SAGE handbook of grounded theory. London, UK: Sage. Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis. London, UK: Sage. Creswell, J. W., & Miller, D. L. (2000). Determining validity in qualitative inquiry. Theory Into Practice, 39(3), 124–130. doi:10.1207/s15430421tip3903_2 De Zeeuw Wright, M. (2010). The dance in contexts: Exploring the complexity of the helping/ healing process with a focus on client satisfaction (Unpublished doctoral dissertation). Wilfrid Laurier University, Waterloo, Ontario, Canada. Dellgran, P., & Hojer, A. (2001). Mainstream is contextual: Swedish social work research dissertations and theses. Social Work Research, 25(4), 243–252. doi:10.1093/swr/25.4.243 Dergal, J. (2011). Family members’ use of private companions in nursing homes: A mixed methods study (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Dill, K. A. (2010). “Fitting a square peg into a round hole”: Understanding kinship care outside of the foster care paradigm (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Doiron, N. H. (2007). People with environmental sensitivities: Life, identity, and services (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Drisko, J. (1997). Strengthening qualitative studies and reports: Standards to enhance academic integrity. Journal of Social Work Education, 33, 1–13. Dunne, C. (2011). The place of the literature review in grounded theory research. International Journal of Social Research Methodology, 14, 111–124. doi:10.1080/ 13645579.2010.494930 Erlandson, D., Harris, E., Skipper, B., & Allen, S. (1993). Doing naturalistic inquiry: A guide to methods. London, UK: Sage. Forchuk, C., & Roberts, J. (1993). How to critique qualitative research articles. Canadian Journal of Nursing Research, 25(4), 47–56. Geertz, C. (1976). The interpretation of cultures: Selected essays by Clifford Geertz. New York, NY: Basic Books. Downloaded by [99.252.168.12] at 11:50 09 November 2017 JOURNAL OF TEACHING IN SOCIAL WORK 19 Glaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory. Mill Valley, CA: Sociology Press. Glaser, B. G. (2002). Conceptualization: On theory and theorizing using grounded theory. International Journal of Qualitative Methods, 1(2), 23–38. doi:10.1177/160940690200100203 Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine. Gowthorpe, J. L. (2010). Making the adjustment: A grounded theory study of what works and does not work in postdivorce parenting relationships (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Gringeri, C., Barusch, A., & Cambron, C. (2013). Examining foundations of qualitative research: A review of social work dissertations. Journal of Social Work Education, 49, 760–773. Guest, G., Bruce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18, 59–82. doi:10.1177/ 1525822X05279903 Helpard, E. J. (2010). Making room to practice: A grounded theory of counsellors’ experiences of unplanned endings (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Jenney, A. C. (2011). Doing the right thing: Negotiating risk and safety in child protection work with domestic violence cases (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Jones, J., & Hunter, D. (1995). Qualitative research: Consensus methods for medical and health services research. British Medical Journal, 311, 376–380. doi:10.1136/bmj.311.7001.376 Krefting, L. (1991). Rigor in qualitative research: The assessment of trustworthiness. The American Journal of Occupational Therapy.: Official Publication of the American Occupational Therapy Association, 45, 214–222. doi:10.5014/ajot.45.3.214 Lach, L. M. (2004). Social experiences of children and adolescents diagnosed with intractable epilepsy: Maternal representations (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Lawson, J. J. (2009). A grounded theory study: Exploring the process of forgiving within significant relationships (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Levy, J. (2004). Deviance and social control among Haredi adolescent males (Unpublished doctoral dissertation). McGill University, Montreal, Québec, Canada. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage. Maynard, B. R., Vaughn, M. G., & Sarteschi, C. M. (2014a). The empirical status of social work dissertation research: Characteristics, trends and implications for the field. British Journal of Social Work, 44, 267–289. doi:10.1093/bjsw/bcs123 Maynard, B. R., Vaughn, M. G., Sarteschi, C. M., & Berglund, A. H. (2014b). Social work dissertation research: Contributing to scholarly discourse or the file drawer. British Journal of Social Work, 44, 1045–1062. doi:10.1093/bjsw/bcs172 McKibbon, K., & Gadd, C. (2004). A quantitative analysis of qualitative studies in clinical journals for the 2000 publishing year. BMC Medical Informatics and Decision Making, 4, 11–20. doi:10.1186/1472-6947-4-11 McLaughlin, A. M. (2006). Clinical social work and social justice (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. McNeill, T. (2001). Holistic fatherhood: A grounded theory approach to understanding fathers of children with juvenile rheumatoid arthritis (JRA) (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Downloaded by [99.252.168.12] at 11:50 09 November 2017 20 M. BRAGANZA ET AL. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. Thousand Oaks, CA: Sage. Milliken, E. J. (2008). Towards cultural safety: An exploration of the concept for social work education with Canadian Aboriginal peoples (Unpublished doctoral dissertation). Memorial University, St. John’s, Newfoundland, Canada. Monahan, M. F. (2001). An exploration of the context of women’s experience of unwanted sex in couple partnerships (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Munoz, M. (2011). Continuum of success: A case study of Colombian refugee women in Canada (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Muskat, B. L. (2008). Identity formation in early adolescents with learning disabilities (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Nagel, D. A., Burns, V. F., Tilley, C., & Aubin, D. (2015). When novice researchers adopt constructivist grounded theory: Navigating less travelled paradignmatic methodological paths in PhD dissertation work. International Journal of Doctoral Studies, 10, 365–383. Norlyk, A., & Harder, I. (2010). What makes a phenomenological study phenomenological? An analysis of peer-reviewed empirical nursing studies. Qualitative Health Research, 20, 420–431. doi:10.1177/1049732309357435 O’Reilly, M., & Parker, N. (2012). “Unsatisfactory saturation”: A critical exploration of the notion of saturated sample sizes in qualitative research. Qualitative Research, 13, 190–197. doi:10.1177/1468794112446106 Oktay, J. S. (2012). Grounded theory. New York, NY: Oxford University Press. Padgett, D. K. (2008). Qualitative methods in social work research (2nd ed.). Thousand Oaks, CA: Sage. Patton, M. Q. (1999). Enhancing the quality and credibility of qualitative analysis. Health Services Research, 34(5), 1189–1208. Reid, A., & Gough, S. (2000). Guidelines for reporting and evaluating qualitative research: What are the alternatives? Environmental Education Research, 6(1), 59–91. doi:10.1080/ 135046200110494 Riedel Bowers, N. (2001). A journey within a journey: A naturalistic study of the early relationship development process in non-directive play therapy (Unpublished doctoral dissertation). Wilfrid Laurier University, Waterloo, Ontario, Canada. Rothwell, D. W., Lach, L., & Blumenthal, A. (2013). Social work doctoral scholarship in Canada (Version V2). Cambridge, MA: Harvard Dataverse Network. Retrieved from http://hdl.handle.net/1902.1/21789 Rothwell, D. W., Lach, L., Blumenthal, A., & Akesson, B. (2015). A scoping review of doctoral scholarship in Canada: Implications for the discipline. Journal of Teaching in Social Work, 35(1–2), 46–64. doi:10.1080/08841233.2014.977988 Rubin, A. (2000). Standards for rigor in qualitative inquiry. Research on Social Work Practice, 10, 173–178. Ryle, G. (1949). The concept of mind. London, UK: Hutchinson. Saini, M. (2011). Qualitative research quality checklist (Unpublished manuscript). Toronto, Ontario, Canada. Saini, M., & Shlonsky, A. (2012). Systematic synthesis of qualitative research. Oxford, UK: Oxford University Press. Scourfield, J., & Maxwell, N. (2010). Social work doctoral students in the UK: A web-based survey and search of the index to theses. British Journal of Social Work, 40, 548–566. doi:10.1093/bjsw/bcn165 Downloaded by [99.252.168.12] at 11:50 09 November 2017 JOURNAL OF TEACHING IN SOCIAL WORK 21 Shenton, A. K. (2004). Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22, 63–75. doi:10.3233/EFI-2004-22201 Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. London, UK: Sage. Strauss, A., & Corbin, J. (1994). Grounded theory methodology. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 217–285). Thousand Oaks, CA: Sage. Timoshinka, N. (2008). Non-hierarchical organizing and international women’s NGOs: An exploratory study (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Tobin, G. A., & Begley, C. M. (2004). Methodological rigour within a qualitative framework. Journal of Advanced Nursing, 48, 388–396. doi:10.1111/jan.2004.48.issue-4 Tracy, S. J. (2010). Qualitative quality: Eight “big-tent” criteria for excellent qualitative research. Qualitative Inquiry, 16, 837–851. doi:10.1177/1077800410383121 Walker, D. E. (2005). Learning to be a social worker: From the voices of students (Unpublished doctoral dissertation). University of Calgary, Calgary, Alberta, Canada. Whittemore, R., Chase, S. K., & Mandle, C. L. (2001). Validity in qualitative research. Qualitative Health Research, 11, 522–537. doi:10.1177/104973201129119299 Worthington, C. A. (2001). Being tested: An investigation of recipient perspectives on HIV testing services using a multi-method approach (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Yan, M. C. (2002). A grounded theory study on cultural and social workers: Towards a dialectical model of cross-cultural social work (Unpublished doctoral dissertation). University of Toronto, Toronto, Ontario, Canada. Yanow, D., & Schwartz-Shea, P. (Eds.). (2006). Interpretation and method: Empirical research methods and the interpretive turn. Armonk, NY: M.E. Sharpe. Zuk, G. (2009). Honoring the subjective: An exploration of the self-reflexive portfolio in social work education (Unpublished doctoral dissertation). Memorial University, St. John’s, Newfoundland, Canada.