5541 Jpa Vol 45 No3_article 3_pg435-455

THEORY DEVELOPMENT: ENHANCING THE QUALITY OF THE CASE STUDY AS RESEARCH STRATEGY IN QUALITATIVE RESEARCH E. Schurink Department of Public Governance Sanlam Centre for Public Management and Governance University of Johannesburg C.J. Auriacombe Department of Public Governance University of Johannesburg

ABSTRACT

G

ood quality research is essential. However ensuring the quality of qualitative research is not as easy as it sounds. An important question in qualitative and evaluation research is centred on the question of how the credibility of the research could be enhanced? This is especially true in research aimed at theory development when a case study strategy is utilised. This is the main theme of this article. Utilising criteria for quality research is one way of doing it. However, as pointed out in this article, due to the different ontological and epistemological beliefs of researchers belonging to the various paradigms, the criteria for trustworthy, credible research can never meet everyone’s approval. Therefore a growing number of scholars (Schwandt 1996) have moved beyond measuring the quality of research by using criteria to implement specific strategies for managing the research process. In general these strategies are also applicable to case studies. However, because the case study strategy is intertwined with theory development, it is reasoned that there are specific formalised strategies aimed at generating theory that could be used to enhance of the quality of case study research. It is also argued that strategies using both an inductive and deductive approach to theory development could be used to construct a more comprehensive methodology for developing theory in case study research.

E. Schurink & C.J. Auriacombe

435

Article INTRODUCTION

ven a brief overview of the available literature on qualitative research is bound to reveal the following:

Firstly, qualitative research is not a new approach for social study. Spanning many decades, it has witnessed various changes throughout the years. Furthermore, it has encircled various traditions and practices with a number of exponents. Notably, qualitative research has used various approaches and methods of collecting, analysing and interpreting data, as well as writing qualitative reports. Secondly, qualitative research is constantly changing. It is clear that as we strive to make sense of the social world and create new knowledge or revisit what we know we tend to use approaches spanning various disciplines in the social sciences. Furthermore, we are increasingly relying on humanities-based approaches and practices. Shank (2006:218) writes: “One thing is for sure – qualitative research will continue to change. Qualitative research has been in motion ever since it first started coming together as a cohesive movement, and there is no reason to suppose that it will not continue to grow and develop”. Thirdly, qualitative research is an umbrella term for different approaches to this type of research. Notably, each has its own theoretical background, methodological principles and aims (Flick, 2007:6). Denzin and Lincoln (2005:xv): write: “There is no one way to do interpretive, qualitative inquiry. We are all interpretive bricolage stuck in the present working against the past as we move into a politically charged and challenging future (emphasis in original)”. Finally, in light of the ever-changing nature of this field of study, there is no uniform definition for qualitative research. As Denzin and Lincoln (2005:xv) state: “The openended nature of the qualitative research project leads to a perpetual resistance against attempts to impose a single, umbrella paradigm over the entire project”. Nevertheless, for our purposes it is useful to have some understanding of qualitative research of which arguably the most realistic is taken from the second edition of The Handbook of Qualitative Research (2000): “Qualitative research is an interdisciplinary, transdisciplinary, and sometimes counterdisciplinary field. It crosscuts the humanities, the social sciences, and the physical sciences. Qualitative research is many things at the same time. It is multiparadigmatic in focus. Its practitioners are sensitive to the value of the multimethod approach. They are committed to the naturalistic perspective and to the interpretive understanding of human experience. At the same time, the field is inherently political and shaped by multiple ethical and political allegiances. Qualitative research embraces two tensions at the same time. On the one hand, it is drawn to a broad, interpretive, postexperimental, postmodern, feminist, and critical sensibility. On the other hand, it is shaped to more narrowly defined positivist, postpositivist, humanistic, and naturalistic conceptions of human experience and its analysis” (Denzin & Lincoln, 2000:1048). Case studies are used when investigators have little control over events and when the focus is on contemporary phenomena within a real-life context – especially when the boundaries between phenomenon and context are not clearly evident. Qualitative case

436

Journal of Public Administration • Vol 45 no 3 • September 2010

studies use documents, physical artefacts, document analysis, interviews, critical incidents focus groups, informal discussion, participant observation and direct observation and quantitative case studies use mailed or online questionnaires/survey post as sources of evidence. Importantly, this type of research seeks to present empirical evidence fairly and rigorously. Sources that may already exist – if relevant to the case – include statistics, archival records, written reports/minutes of events and activities, discussion papers/ publications, press photos and cuttings, bulletins/newsletters and analysis of specific results. Case studies could be done without leaving the library and the telephone, unlike participant observation – which invests a lot of time in field work – and ethnographic studies – which require a lot of time in the field and emphasise observational evidence (cf. Yin, 1994:1-12, Yin 2003: 85-96 and Simons, 2009:164-7). Before turning to a more in-depth discussion of the quality of case studies, it is necessary to address the different, often-conflicting viewpoints of the concept “case study”.

CASE STUDIES EXPLAINED chram (2006) emphasises that the case study strategy differs from other methodological frameworks in that it is more of a framework for designing a study than for conducting study. Since its inception in the 1920’s, researchers who have used the case study in qualitative research have been unable to present particular data gathering techniques for case study research. This becomes clear by reviewing the case study’s ability to adapt to all the theoretical approaches and methodological frameworks, such as life history, phenomenology, grounded theory and ethnographic research. However, whether one considers the case study as a way of conceptualising human behaviour or merely as a means of encapsulating it, its strategic value lies in its ability to draw attention to what can be learned from the single case (Schram, 2006). “This qualitative case study is an approach to research that facilitates exploration of a phenomenon within its context using a variety of data sources. This ensures that the issue is not explored through one lens, but rather a variety of lenses which allows for multiple facets of the phenomenon to be revealed and understood” (Baxter & Jack, 2008). A case study is an in-depth study (usually longitudinal) of one or a few cases, in contrast to a more superficial cross-sectional study of a larger sample (Bailey, 1978:42). According to Creswell (2007) a case study can be regarded as an exploration or in-depth analysis of a “bounded system” (bounded by time and/or place), or a multiple or single case, over a period of time. The case being studied may refer to a process, activity, event, programme, individual or multiple individuals. It might even refer to a period of time – rather than a particular group of people. Thus, a case is “…a unit of human activity embedded in the real world” (Graham, 2000:1). The researcher explores a single entity or phenomenon, “the case” bounded by time and activity (a programme, event, process, institution or social group) and collects detailed information by using a variety of data collection procedures during a sustained period of time (Yin, 1994, 2003). The researcher situates this system or case within its larger context, but the focus remains on either the case or the issue it illustrates (Creswell 2007). E. Schurink & C.J. Auriacombe

437

Case studies in qualitative research emphasise the detailed contextual analysis of a limited number of events or conditions and their relationships. Stake (1995) emphasises that the most important advantage of a case study is that it enables the researcher to present the complexity and multi-dimensionality of a case. Therefore, the case study as a qualitative research strategy is best applied to understanding social phenomena in relation to its wider context (Stake, 1995). Unlike quantitative research the wider context of the case is not used to explain, compare and verify. In similar vein, it does not seek to identify the causes of phenomena, but rather seeks to gain a better understanding and to address the complexity of the case (David 2006:xxvii). In terms of Yin’s context, the literature analysis will serve to look outside the firm boundaries and enable comparison with others. This is in terms of both practices and performances to enable the process of acquiring external explicit and tacit knowledge. Such newly-acquired knowledge, once integrated with previous internal knowledge of the critical case, creates a new knowledge base that may give rise to certain improvements and innovations (cf. Yin 1994, 2003). Merriam (1998) argues that the case study is designed to gain an in-dept understanding of the situation and the meaning involved. The interest is in the process rather than the outcomes; in context rather than specific variables; in discovery rather than information. In quantitative research, case studies seek to explain the causal links in real life situations that are too complex for experimental or survey strategies to do (Yin 1994:15). Therefore, the case is not selected for its unique characteristics. Rather, it is chosen through scientific acts of selection and exclusion (sampling), to be compared with other similar or different cases. In this sense the case study is used to explain an intervention and the real-life context in which it occurs (David 2009:xxvi). Clearly, within the qualitative methodology the “core strength of case method lies in exploration and description” (David, 2009:xxvi). Yin (1994) distinguishes between four types of case study designs: A holistic singlecase; an embedded single-case; a holistic multiple-case, as well as an embedded multiple-case. Multiple-cases are usually chosen to capture the variety among cases. In terms of the difference between holistic versus embedded approaches, Yin (1994) states that the difference depends upon the type of phenomenon under study. Embedded analysis focuses on sub-units, therefore, embedded units, such as the Department of Public Governance as a sub-unit of the University of Johannesburg. When the case study focuses solely on the global nature of an institution or programme a holistic case method will be conducted. A weakness of the embedded technique is that it is overly focused on the sub-unit level. Therefore it may fail to go back to larger aspects of the analysis. In turn, a weakness of the holistic technique is that it may risk being conducted at an abstract level, missing clear measures of data. Three types of case studies with different purposes can be identified (Schram 2006): • The intrinsic case study solely focuses on the aim of gaining a better understanding of the particular case. • The instrumental case study is used to provide insight into or elaborate on a theory, as well as to gain a better understanding of a social issue through studying the case. This case study merely serves the purpose of facilitating the researcher’s gaining of knowledge about a social issue.

438

Journal of Public Administration • Vol 45 no 3 • September 2010

• The collective case study is instrumental in nature and is extended to a number of cases. The researcher focuses on furthering his/her understanding about a general phenomenon or condition. Cases are chosen so that comparisons can be made between cases and concepts. In this way, theories can be extended and validated. Case study research needs to be carefully planned and designed. The following questions may be helpful when designing a case study (cf. Simons 2009): • How would you constitute your case? For example what is your unit of analysis (a classroom, an institution, a programme)? • Where do the boundaries of your case lie? (They might change; the event is sometimes only decided at the end) • What audiences do you hope to address? • What key questions (for example evaluation questions) or foreshadowed issues may guide your collection of data? What is the rationale/theory for each? • What is your criterion for selection of the case or cases – intrinsic, instrumental, other region or acceptability? • What ethical norms/guidelines will underpin your data collection and analysis and how would you do this? • Do you want to engage participants in the identification of issues, data gathering and analysis and how would you do this? • Do you wish to start with a theoretical framework or generate a theory of the case? • Whose values matter – participants, stakeholders, institution? • Which methods (the most appropriate) are you going to use to gain an understanding of the issues? Some researchers suggest that one of the problems of case study research is that one cannot generalise. In fact, this is not the case. However, the way in which one generalises from a case differs from adopted forms of research that utilise large samples and quantitative measures. There are two particular ways to approach this issue. One is to examine how we can generalise from the case to another situation of a similar or dissimilar nature. The other is to find out what we can learn more directly from the case itself. All these forms of generalisation are not formal propositional generalisations. Moreover, all maintain a connection with the context in which they were generated. The following forms (cf. Simons, 2009:164-7 and Flyvbjerg, 2006:224-228)) of generalising from the single case are highlighted: Cross-case generalisation – is adopted in a collective or multi-site case study where common issues are identified across cases and interconnecting themes. There is a degree of abstraction and it is the closest to propositional form. Naturalistic generalisation – recognises similarities and differences to cases with which readers are familiar, appeals to tacit knowledge, provides vicarious experience and is useful in policy context and professional practice. Concept generalisation – identifies a concept theorised in the case, which has significance in other cases – whether similar or dissimilar. E. Schurink & C.J. Auriacombe

439

Process generalisation – here it is the process, not the content or context that is applicable to a wide variety of contexts. Situated generalisation – relies on retaining a connectedness with the situation in which the understanding first evolved to ensure that it remains general in nature. It is enhanced in practice contexts by a climate of trust and confidence obtained through shared research experience. Finally, in-depth particularisation also provides insights that are developed through in-depth understanding of the particular and aspires to capture the essence of the specific way we all recognise phenomena. Therefore, by studying the uniqueness of the case in all its particularity, we come to understand the universal (Simons, 2009).

CRITERIA TO ENHANCE THE QUALITY OF CASE STUDIES IN QUALITATIVE RESEARCH n general, researchers will agree that social science research is a systematic rigorous and organised process, whereby knowledge (evidence) is gained to provide an accurate or truthful representation of a phenomenon. However, agreeing on this does not mean that researchers will agree on what methods to be followed to produce truthful research (Mouton & Marais, 1996:192). In fact, there is a vast difference between quantitative and qualitative researchers – and even among qualitative researchers – as to how the social world should be studied. Quantitative researchers believe that human behaviour can be explained from the outside (epic), by means of objectivistic observation. They believe in using general scientific laws (erklaren), which demand that good research should have: (i) internal validity, (ii) external validity or representativeness, (iii) reliability and (iv) objectivity. On the other hand, qualitative researchers believe that human behaviour can only be understood from an insider’s point of view (emic), by gaining insight into the meaning (verstehen) that the participants attach to his/her life world. They generally regard these constructs as inappropriate in establishing the “truth value” of a qualitative research project. Some qualitative researchers (Schwandt, 2007) have even started questioning the alternative criteria of trustworthiness that Lincoln and Guba (1985) developed. Many attempts have been made to develop criteria for good qualitative research1, which is beyond the scope of this article. Cohen and Crabtree (2008) undertook an analysis of the literature on criteria for good qualitative research and found no less than 29 journal articles and 16 books or book chapters that offered criteria for evaluating the quality of qualitative research. Due to the range of theoretical frameworks and methodological approaches guiding data collection and analysis embedded in the different paradigms of qualitative research, no consensus on the criteria to be used to judge the quality of case studies has been reached. Judging from the above, it is no easy task to enhance the trustworthiness of case study research. From a scan of the recent literature (Schwandt, 2007), it appears that presently at least two distinct perspectives on assessing the quality of qualitative research are found amongst scholars. These are: (i) qualitative and quantitative research should be

440

Journal of Public Administration • Vol 45 no 3 • September 2010

evaluated by the same measures (ii) qualitative research should be evaluated by standards that have been particularly developed for it, and (iii) what Holloway and Wheeler (2002) call criteriology. That is, abandoning measures to assess qualitative work and bringing in measures to ensure rigour and quality in the research process (Schwandt, 2007). We will now take a look at each of these viewpoints.

Qualitative and quantitative research should be evaluated by the same measures Here, there is the belief that one set of criteria should be applied to all scientific research. Thus, four standard criteria are used for both qualitative and quantitative research: internal validity, external validity, reliability and objectivity. Using these criteria in qualitative research is in line with the positivist paradigm.

Qualitative research should be evaluated by standards that have been particularly developed for it Post-positivists or modernists assess the quality of study with regards to its ability to: (a) generate theory; (b) be empirically grounded and scientifically credible; (c) produce findings that can be generalised; (d) be internally reflexive in terms of taking account the researcher’s research strategy on the findings that have been produced. Many attempts have been made to develop criteria for “good” qualitative research within this modernistic framework (Giorgi, 1988; Kavale, 1996; and Stiles, 1993). The alternative constructs proposed by Lincoln and Guba (1985), are among the best-know attempts. According to Lincoln and Guba (1985) the following are four alternative constructs that reflect the assumptions of the qualitative paradigm more accurately:

Credibility/authenticity versus internal validity Is there a match between research participants’ views and researchers’ reconstruction and representation of it? Here the goal is to demonstrate that the research was conducted in such a manner as to ensure that the subject of investigation was accurately identified and described and is credible to those who constructed the original realities. Credibility emphasises striving for truth through the qualitative research process. Therefore, plausibility and accuracy would be an important dimension of good qualitative research.

Transferability versus external validity Can the findings of the research be transferred from a specific situation or case to another? “A qualitative study’s transferability or generalisability to other settings may be problematic. The generalisations of qualitative findings to other populations, settings, and treatment arrangements – that is, its external validity – is seen by traditional canons as a weakness in the approach” (Marshall and Rossman, 1995:144). The researcher should provide sufficient descriptive data to allow the reader to evaluate the applicability of the data to other contexts. Researchers can strengthen their E. Schurink & C.J. Auriacombe

441

studies’ usefulness for other settings substantially by designing their studies in multiple cases, multiple informants and more than one method of collecting data.

Dependability versus reliability Dependability refers to stability of data over time and in different conditions. The question here is: Is the research process logical, well documented and audited?The researcher uses this construct to account for changing conditions in the phenomenon under study, as well as for changes in the study’s design by increasing refined understanding of the setting. Marshall and Rossman (1995:145) point out that dependability represents a set of assumptions that differs greatly from those shaping the positivistic construct of reliability: “Positivist notions of reliability assume an unchanging universe where inquiry could, quite logically, be replicated. This assumption of an unchanging social world is in direct contrast to the qualitative/interpretive assumption that the social world is always being constructed, and the concept of replication is itself problematic”. Steps should be taken to establish dependability. These steps should be built into the research process to repeat and affirm researchers’ observations.

Confirmability versus objectivity Does the researcher provide evidence that corroborates the findings and interpretations by means of an internal audit? This construct captures the traditional concept of objectivity. This implies that one should ask whether the study findings could be confirmed by another study. In doing so, evaluation is moved from some inherent characteristic of the researcher (objectivity) and is placed on the data. The appropriate criterion for qualitative research is thus: Does the data help to confirm the general findings and lead to their implications?

MEASURES TO ENSURE RIGOUR AND QUALITY IN THE RESEARCH PROCESS onstructivists argue for quality criteria that translate internal and external validity, reliability and objectivity into trustworthiness and authenticity. Verifying the credibility or authenticity of the research would therefore be a process negotiated between researchers and readers. Researchers are responsible for reporting information (for example data excerpts, how the researcher dealt with tacit knowledge and information about the interpretive process), so that readers can verify the data, its analysis and interpretation, thus deciding whether the study is in fact credible (Cohen & Crabtree, 2008). Subjectivists or constructionists would mostly agree with Schwandt (2007) that all measures to assess qualitative research should be abandoned. Constructivists emphasise the fact that qualitative researchers are engaging themselves with difficult questions and issues in the reasoning process through which they arrive at their answers. Therefore, it is not only important to properly plan their studies, but also to provide a reflexive account of how it was done, for example constructing natural histories, research stories, or internal audits, and/or defending the logic of their decisions. It is a good idea to keep research diary from the onset. This will help the researcher to refer

442

Journal of Public Administration • Vol 45 no 3 • September 2010

to particular issues s/he is confronted with during the various developmental phases of the research proposal, as well as the decisions taken during the research process and the reasons for these decisions It is important to note that the post-modern trend in qualitative research encourages researchers to undertake and record a continuous critical analysis of their thinking and feelings concerning their conceptual framework, research questions; methods; values, biases; awareness of their presence in the very situations they want to study, as well as the inter-subjective dynamics between their role as researchers and the research they are conducting. Such a writing style forms an important tool that can be used to enhance the quality of the research. From the above, it becomes clear that the researcher should strive to meet certain criteria that will help ensure quality research. However, this is not as easy as it sounds. The main reason is that qualitative research is grounded in a range of philosophical and theoretical frameworks and uses a variety of methodological approaches to guide data collection and analysis (Cohen & Crabtree, 2008). Since a clear and universal solution to the quality question or criteria for good qualitative research has to date not been attainable, researchers started to move from applying criteria, to implementing strategies to promote quality in the research process. Subsequently, the notion of managing the research process became an important notion to help ensure quality research.

GENERAL STRATEGIES TO ENHANCE THE QUALITY OF CASE STUDIES IN QUALITATIVE RESEARCH espite the diversity range of criteria discussed above, as well as the corresponding literature (Cohen & Crabtree 2008), it is widely believed that the quality of any scientific study rests on its credibility. Lincoln and Guba (1985) supported this belief and regarded this as the most significant characteristic of quality research. Furthermore, the authors outlined various strategies to increase the credibility of qualitative research. These include: Prolonged engagement and persistent observation in the field will increase the likelihood that credible results will be produced Scientific rigour is valued in research circles as it is associated with the quality of research outcomes. As highlighted in the above section research is critiqued to judge the rigorousness of the study. One of the main criticisms launched against qualitative research (including case studies) is that it lacks rigour. However, as discussed above, the criteria used to judge the rigour of qualitative studies are based on rules developed to judge quantitative studies. However, the outcomes of qualitative and quantitative studies differ vastly. Therefore, rigour needs to be defined differently for qualitative research, since the desired outcome is different (interpretation versus explanation). In quantitative research, rigour is reflected in conciseness and objectivity and leads to a rigid adherence to research designs and precise analyses. On the one hand, rigour in qualitative research requires the researcher to adhere to openness and flexibility, and on the other hand to scrupulously follow a philosophical perspective and a thoroughness and rigorousness in E. Schurink & C.J. Auriacombe

443

data collection and analysis (http://www.fortunecity.com/greenfield/grizzly/432/rra3.htm). As Gummesson (2007:4) points out: “Case study research tries to respond to complexity by providing rich and thick descriptions in the sense suggested by Geertz (1973). But the genome of case study research stretches beyond such descriptions. It allows the study of complexity, context, ambiguity and chaos. It allows a holistic, systemic approach with an unlimited number of variables and links”. Comprehensive field notes could help researchers to undertake, generate and analyse data in a systematic and rigorous fashion (Schwandt, 2007:116).

Triangulation Triangulation is an activity where more than one methodological approach is brought to bear on a single point with the aim to enhance scientific rigour (Denzin, 1989:235). Researchers using the case study strategy could therefore systematically triangulate more than one case (the collective case study); investigator; paradigm (mixed methods) theory and methods of data gathering and analysis with the aim of enhancing the scientific rigour of their studies (Flick, 2007:42-44).

Peer debriefing This involves exposing oneself to a disinterested peer in an analytical session and using the peer as a sounding board for new ideas, insights and hunches gained during the research process.

Member checks Member checks are performed by “testing” the data, analytic categories, interpretations and conclusions with members of those groups from whom the data was originally obtained. This can be done both formally and informally during the course of observation and conversation (Cohen & Crabtree, 2008).

External auditing With this strategy a researcher who is not involved in the research process examines “both the process and product of the research study. The purpose is to evaluate the accuracy and evaluate whether the findings, interpretations, and conclusions are supported by the data” (Cohen & Crabtree, 2008).

THE ROLE OF THEORY IN ENHANCING QUALITY IN CASE STUDY RESEARCH particular strength of case studies is its adaptability. This becomes clear when looking at the case study’s ability to adapt to all the theoretical approaches and methodological frameworks, such as life history, phenomenology, grounded theory and ethnographic

444

Journal of Public Administration • Vol 45 no 3 • September 2010

research. However, embedded in this strong point is also its greatest weakness, namely an inability to develop its own methodology or research strategy. Therefore, according to the opinions of scholars such as Schram (2007), the case study strategy is more of a framework for designing a study than conducting the study. Therefore, it should be considered in a serious light. Gummesson (2007) points out that case study research has not necessarily changed during the past 20 years and that serious changes are needed to enhance the quality of case study research and to establish a proper case study methodology. When considering this, the question is whether specific formalised research strategies aimed at enhancing the quality of case study research, such as grounded theory and analytical induction, could not be used to develop a framework for a case study methodology. To further the viewpoint, we will now turn to a discussion of the two most prominent research strategies whereby deductive and inductive theory can be developed and tested in qualitative research.

The role of philosophical beliefs There exist alternative answers to each foundational question. Different beliefs of ontology (how one sees reality), and epistemology (how the researcher thinks social phenomena should best be studied) will influence the way that s/he will go about conducting the research. It is therefore important to get clarity on one’s own world-view and one’s belief of how things should be studied. Researchers bring their own specific beliefs to a particular study. This often includes training in a particular field, knowledge of substantive topics, a particular standpoint, and theoretical approaches or a conceptual framework. Depending on their orientation, researchers have an idea or way of reasoning on how the study should proceed in order to answer the research question as truthfully as possible.

Develop a conceptual framework for the study. A conceptual framework is an intermediate theory that attempts to connect all aspects of inquiry (for example, problem definition, purpose, literature review, methodology, data collection and analysis) into one framework. The use of a conceptual framework helps us to make sense of occurrences. Furthermore, it ensures that important issues are not overlooked and provides a set of constructs to be investigated while guiding our interpretation and focus. This implies that the research questions should clearly relate to the research approach and the research strategy to be followed. The research setting (sample) chosen and the methods of data collection and analysis must enable the researcher to answer the research question(s) and to deal with possible threats related to research (Maxwell, 2005). Conceptual frameworks are like maps that give guidance to empirical inquiry and analysis (Baxter & Jack 2008:553). Therefore, a conceptual framework becomes a researcher’s first attempt at making some explicit theoretical statements about how the research question(s) relate to prior research and theory; to our own experience; and to the way we are going to address the research problem (Pare, 2002). According to Miles E. Schurink & C.J. Auriacombe

445

and Huberman (1994:18), the conceptual framework serves several purposes, namely to provide a boundary to the study (identifying who will and will not be included in the study); show the relationships that may be present based on logic, theory and/or experience, as well as give the researcher the opportunity to categorise constructs. According to Baxter and Jack (2008:553) “one of the drawbacks of a conceptual framework is that it may limit the inductive approach when exploring a phenomenon.” They suggest that, in order to overcome this disadvantage, researchers should constantly reflect on their own research decisions and use peer debriefing to determine if their thinking has become too driven by the framework. It should be remembered that, since the common aim of case study research is typified as theory building and not theory testing, the conceptual framework and its constructs could be used solely as a starting point in the research process (Baxter & Jack, 2008:553). Yin (1999) is of the opinion that even exploratory case study research should use a conceptual framework to define the priorities to be explored.

Constantly and systematically reflect on all the research issues encountered and the decisions taken Any factors that may have affected data collection, analysis and interpretation should be continuously reflected upon by means of field notes that are written up in a natural history format. Many researchers (Denzin, 1997, 1995, Hammersley & Atkinson, 1995, Etherington, 2006) have extensively discussed the problem of reflexivity. Reflexivity implies a critical self-examination of our roles as researchers throughout the entire research process. The quality of qualitative research could be greatly-enhanced by undertaking and recording a continuous critical analysis of our thinking and feelings concerning our: i) conceptual framework; ii) research questions; iii) methods; iv) values and biases; v) awareness of our presence in the situations we want to study; as well as vi) the inter-subjective dynamics between ourselves as researchers and the research we are conducting. In short, reflective remarks can help one to start thinking, making deeper and more general sense of what was happening, as well as explaining things in a conceptually coherent way (Paré, 2002). Qualitative researchers establish the credibility, dependability and confirmability of their findings by keeping a diary or journal as a traceable audit trail. This will show that the research was done in a rigorous fashion (Seale, 1999). Reflexivity could be seen as the researcher’s attempts to find the meaning of his/her research journey. The analysis of qualitative data is mostly regarded as a mechanical process of sorting, organising and indexing the information obtained. Reflexivity could help the researcher to be ontologically, epistemologically, personally, theoretically and methodologically interconnected in the research process. This is particularly true of the data analysis phase. Researchers bring their own ontological, epistemological and theoretical assumptions, as well as their own backgrounds, personal, emotional and interpersonal influences to the research process (Mauthner & Doucet, 2003). By admitting this, critically reflecting on these issues, as well as discussing it openly and honestly, the credibility of the research is greatly enhanced.

446

Journal of Public Administration • Vol 45 no 3 • September 2010

Therefore, it has become more important for qualitative researchers to reflect on the research process in terms of: What happened? Who was involved? Where did the activities occur? What circumstances or issues had an impact on the data? What are the major issues emerging? What issues need to be followed up? It also involves further exploring the meaning of the data by asking “so what” and “what if” questions. The reflexive process is very helpful to relating occurrences, such as words, expressions, interactions, social processes, events and other occurrences, as well as values and norms of particular groups of people. Discovering such linkages is important in selecting further theoretical incidents, persons, behaviours, etc. and in establishing and verifying evolving ideas, themes and typologies.

Focus the research clearly All research originates from a research idea. This idea or a notion focuses on what to study. The real challenge, however, is to take an idea and transform it into a research problem. To do this, the researcher must have an idea of the present stance of knowledge on the study topic. The gaps in the research need to be identified, as well as how and where the proposed study will fit into the present research. Importantly, study’s contribution on a theoretical, methodological or practical level needs to be clarified, while the motivation for the study should be outlined. The research question is developed by refining the problem statement. It should be stated in such a way that it provides a well-defined focus to collect data systematically (Baxter & Jack, 2008). “The research question should be defined narrowly enough so that the research is focused and broad enough to allow for flexibility and serendipity” (Pandit, 1996). The purpose of the research will indicate whether the answer to the research question requires any of the following activities: an exploration, description, explanation, evaluation, or a combination of these types. As Pandit (1996) states: “A good source of research questions is the ‘technical literature’ (i.e., reports of research studies and theoretical and philosophical papers characteristic of professional and disciplinary writing) on the general problem area”. The research questions should fit into the conceptual framework. Furthermore, the research question is important in building theory from case studies. This means that research questions should clearly relate to the theoretical approach taken, the research strategy, research setting (sample) chosen, as well as the methods of data collection and analysis. The research design should enable the researcher to answer the research question(s) and to deal with possible threats related to the quality of the research (Maxwell, 2005). Other factors that may influence the research question include: personal interest and expertise, ethical standards, as well as other practical considerations, such as available resources. Mason (1996:15-16) provides the following description of a research question: “A research question is a question which the research is designed to address … and, taken together, your research questions should express the essence of your enquiry. Therefore, you need to have done a great deal of thinking about the essence of your enquiry in the E. Schurink & C.J. Auriacombe

447

sense of its ontology, its epistemology, and most importantly its intellectual puzzle2, in order to be able to formulate research questions sensibly and coherently”.

Establish the boundaries of the case within a conceptual framework Once basic research questions have been generated and the research is focused, the next step is to select the first case. Cases (the principal units of data in this research) should be selected according to the principle of theoretical sampling. Unlike the sampling done in quantitative investigations, theoretical sampling cannot be planned before embarking on a grounded theory study. The specific sampling decisions evolve during the research process itself (Strauss & Corbin, 1990:192). When doing case study research, it is important to establish the boundaries of the case. The decision on these boundaries, as well as what to include and exclude, is conceptually driven by the previous knowledge of the phenomenon under study, the research question and how the question will be answered. Baxter and Jack (2008:547) state that: “A qualitative case study design is similar to the development of inclusion and exclusion criteria for sample selection in a quantitative study. The difference is that these boundaries also indicate the breadth and depth of the study and not simply the sample to be included”. Here it is important to note that the case study’s main aim is to gain a holistic understanding of the overall case and not necessarily to understand the various parts of the case, or the contributing factors that influence the case (Baxter & Jack, 2008:555). Therefore, although the case may differ according to the parts included in it, the focus will be on understanding the case itself. For example, depending on the focus, a study of Local Government can include or exclude service providers, the public, councillors and the public.

Record and manage the data in a systematic way Because case study research generates a large amount of data from multiple sources, it is important to organise the data systematically to prevent the researcher from becoming overwhelmed by the amount of data and from losing sight of the original research purpose and questions. Advance preparation helps the researcher to handle large amounts of data in a documented and systematic fashion. Researchers should prepare databases to help categorise, sort, store, and retrieve data for the purpose of analysis. The formats used for storing data should also enable easy management and systematic retrieval, so that patterns could be discovered. Case study research is flexible, but when changes are made, they should be documented systematically. Field notes and the auditing trail should be used to categorise and reference data, so that it is readily available for interpretation. Researchers should use field notes to record: i) their observations of the physical set-up, what happened, what was heard, seen and experienced; ii) their methodological insights on how to improve the quality of the data gathering process, as well as whom to talk to next; iii) their theoretical hunches reflecting critically on the

448

Journal of Public Administration • Vol 45 no 3 • September 2010

development of new themes from existing categories; and iv) their personal notes that critically reflect on their feelings about the research. In short, field notes document the work in progress. They record testimonies, stories, and illustrations that can be used to analyse the data.

Link the data to the research questions and insights gained Case study research is known to use triangulation in terms of data gathering and analysis methods to enhance the scientific rigour of studies. Potential data sources, such as documentation, archival records, interviews, physical artefacts, direct observations and participant-observation are used to reach a holistic understanding of the phenomenon being studied. Data from these multiple sources are then united in the analysis process – rather than handled individually. As Glaser and Strauss (1967:65) state: “In theoretical sampling, no one kind of data on a category or technique for data collection is necessarily appropriate. Different kinds of data give the analyst different views or vantage points from which to understand a category and to develop its properties; these different views we have called slices of data. While the [researcher] may use one technique of data collection primarily, theoretical sampling for saturation of a category allows a multifaceted investigation, in which there are no limits to the techniques of data collection, the way they are used, or the types of data acquired”. When data is collected for the purposes of generating theory the analyst jointly collects, codes, and analyses of his/her data. Furthermore, the researcher decides what data to collect next and where to find it, in order to develop his theory as it emerges (Glaser and Strauss, 1967:45). Within this general framework, data analysis for each case involves generating concepts through the process of “coding which represents the operations by which data is broken down, conceptualised, and put back together in new ways. It is the central process by which theories are built from data (Strauss & Corbin, 1990:57). Lehmann (2001) describes the inductive generating theory process as a spiral that starts by collecting “slices of data” in a substantive area of enquiry. “Each data source is one piece of the ‘puzzle,’ with each piece contributing to the researchers’ understanding of the whole phenomenon. This convergence adds strength to the findings as the various strands of data are braided together to promote a greater understanding of the case”(Baxter & Jack, 2008: 554). Open coding (coding “without a preconceived set of codes”) (Glaser, 1978) is used to codify and categorise the data in a continuous process. Importantly, it moves toward theoretical saturation (theoretical densification) of concepts represented by a substantive theory. Therefore, data is analysed by extracting a set of categories and their properties generated by means of a constant comparative process. As in any other qualitative study the data collection and analysis occur concurrently. This process lasts from data collection to data analysis and therefore generates emergent themes that the data collection and theory development process.

E. Schurink & C.J. Auriacombe

449

There are three types of coding: open coding, axial coding and selective coding. These are analytic types and the researcher not necessarily moves from open, through axial to selective coding in a strict, consecutive manner. Open coding refers to the part of analysis that deals with the labeling and categorising of phenomena, as indicated by the data. The product of labeling and categorising are concepts and can be regarded as the basic building blocks in grounded theory construction. Open coding requires applying of what is referred to as “the comparative method”. This involves asking questions and making comparisons. The data is initially broken down by asking simple questions, such as what, where, how, when and how much. Subsequently, data is compared and similar incidents are grouped together and given the same conceptual label. The process of grouping concepts at a higher, more abstract level is termed categorising. This means that a method of constant comparison is used when new data is gathered. The subjects’ observed actions and recorded perceptions are constantly compared with that of new subjects or new observations in order to generate theory. Therefore, the researcher constantly asks him/herself: “How does what I already have differ from what I have now found?” The aim of this constant comparison method is to look for similarities and differences in the data. From this process the researcher identifies underlying uniformities in the indicators or incidents (actions, events and perspectives) and produces a coded category or concept. These categories are compared with one another and with new incidents to sharpen the definition of the concept and to look for possible new categories. Importantly, categories are clustered together to form themes. As codes and memos accumulate the researcher starts to see relationships between them. This process is called ”theoretical coding. Theoretical codes “emerging from the process of open coding and theoretical memos, weave a new story from the fragmented codes generated through the open coding process” (http://epress.anu.edu.au/info_systems/mobile_devices/ch05s04.html). Whereas open coding fractures the data into concepts and categories, axial coding reconstructs the data in new ways by making connections between a category and its sub-categories (not between discrete categories as is the case with selective coding). Thus, axial coding refers to the process of developing main categories and their sub-categories. Subsidiary categories relate to the core category according to the paradigm model. The basic purpose is to enable the researcher to think systematically about data and to relate elements rigorously in complex ways. The basic idea is to propose linkages and look to the data for validation (move between asking questions, generating propositions and making comparisons). Selective coding involves the integration of the categories that have been developed to form the initial theoretical framework. Firstly, a storyline is either generated or made explicit. A story is simply a descriptive narrative about the central phenomenon of study and the storyline is the conceptualisation of this story (abstracting). When analysed, the storyline becomes the core category: “The core category must be the sun, standing in orderly systematic relationships to its planets” (Strauss & Corbin, 1990:124).

450

Journal of Public Administration • Vol 45 no 3 • September 2010

The research objective is to generate theory “that accounts for the patterns of behaviour which is relevant for those involved” (Glaser, 1978:93). To achieve this goal the researcher must discover the core category and delimit the investigation around it. The core category is the pivotal point for the theory; most other categories relate to it, and it accounts for most of the variation in pattern and behaviour. The core category has the prime function of integrating the theory and rendering the theory dense and saturated as the relationships increase (Glaser, 1978:93). As a general rule, core categories (those with the greatest explanatory power) should be saturated as comprehensively as possible (Pandit, 1996). The data generates insights, hypotheses and generative questions, which the researcher follows up on through generating further data. This leads to tentative answers to questions and constructing concepts that are verified through further data collection. When the researcher proposes plausible relationships between themes certain patterns and theories emerge. Tentative theories are further explored through additional data and are tested by means of theoretical sampling. At this stage in the process, the role of the literature becomes very important since it provides more data to be compared with existing grounded data. This can raise the theoretical level and lead to improved construct definitions (as suggested by Eisenhardt, [1989]). The criterion for judging when to end theoretical sampling is the categories or theory’s “theoretical saturation”. By this term, Glaser and Strauss refer to the situation in which: “... no additional data is being found whereby the (researcher) can develop properties of the category. As he sees similar instances over and over again, the researcher becomes empirically confident that a category is saturated ... when one category is saturated, nothing remains but to go on to new groups for data on other categories, and attempt to saturate these categories also.” (1967:65.) At this stage, when the theory becomes dense with concepts and enriched by relevant extant literature, the researcher has “discovered” a substantive theory. Substantive theories are applicable to the particular area of empirical enquiry from which they emerged (Glaser and Strauss, 1967). Schwandt (2007:131) emphasises the fact that substantive theory is developed by simultaneously employing techniques of induction, deduction and verification. A distinction is made between substantive and formal theory. A substantive theory is a description and abstraction of what goes on in a particular kind of social setting, for example, a specific municipality. Analytic abstractions are used to discuss such settings. However, no claim is made that the abstractions apply to other situations. In formal theory, abstractions and hypotheses about the relationships among these abstractions are developed. These hypotheses should then explain phenomena in many kinds of settings. Formal theory is concerned with a conceptual area of study, such as organisational theory. The writing of theoretical memos is part and parcel of reflexivity. Writing theoretical memos is especially important in the open coding process. Since the analyst cannot readily keep track of all the categories, properties, hypotheses and generative questions that evolve from the analytical process, there must be a system for doing so. The use of memos constitutes such a system. “Memos are not simply ‘ideas.’ They are involved in the formulation and revision of theory during the research process” (Pandit, 1996). When writing these memos the researcher needs to critically reflect on the major issues E. Schurink & C.J. Auriacombe

451

emerging, as well as the issues that need to be followed up. “At least three types of memos may be distinguished: code memos, theoretical memos and operational memos. Code memos relate to open coding and thus focus on conceptual labelling. Theoretical memos relate to axial and selective coding and thus focus on paradigm features and indications of process. Finally, operational memos contain directions relating to the evolving research design” (Pandit, 1996). According to Glazer (1978:83) memos are “the theorising write up of ideas about codes and their relationships as they strike the analyst while coding” Furthermore, memos raise the theoretical level via a continuous process of comparison and conceptualisation. They also provide freedom, flexibility, and enhance creativity (http://epress.anu.edu.au/ info_systems/mobile_devices/ch05s04.html). It is clear from the above that case study research tries to respond to complexity of reality by exploring and describing complex situations. Its particular strength is that it could provide a multidimensional perspective to help provide a better understanding of the slice of reality being studied (Remenyi et al., 2002:5). However, the thrust of case study research stretches beyond such descriptions. It allows for generating theories explaining the complexity, context and ambiguity of real life (Eisenhardt, 1989). Although generating grounded theory involves an iterative process in which researchers cycle back and forth between generating theory and processing data, it aims to produce substantive theory and not to develop universally-applicable theory (Schwandt, 1993:9). The process developing theory from case study research ends when theoretical saturation is reached (when new cases provide very little in the way of new insights). However, some researchers (Hammersley, 2010) are starting to question whether generating theory is enough. In fact, there is an increased perception among researchers that theory development should begin with developing substantive theory in an inductive manner, and then be broadened to constructing formal theory by implementing a deductive strategy, such as analytic induction (Schwandt, 2007). Furthermore, as Hammersley (2010 at http://www.qualitative-research. net/index.php/fqs/article/view/1400/2994) states: “It is hard to understand how the development of theory can be effective if it specifically avoids any concern with whether or not the emerging theoretical ideas are likely to be true”. The two best-known research strategies to build theory include grounded theory, propagated by Glaser and Strauss (1967), and analytical induction, propagated by Lindesmith (1947). Ideally, the case study method should include both grounded theory and analytic induction strategies that could be used to generate and test theory. By using the formalised strategies advocated by grounded theory and analytical induction, one can move beyond merely relying on criteria and strategies to enhance the quality of case study research.

CONCLUSION ase studies have an important function in generating hypotheses and building theory (cf. e.g. Eisenhardt, 1989). Today scientists agree that case studies could well be used to generate hypotheses, as well as hypothesis testing. A high-quality

452

Journal of Public Administration • Vol 45 no 3 • September 2010

case study therefore implies a comprehensive and intensive study of complex situations. Therefore, case studies could be seen as the analysis of a micro-cosmos of realities that could be used to understand the wider reality. It offers the opportunity to gain a holistic view of an entity and the process of studying it. With above features of case studies as foundation attempts were made to provide specific strategies related to theory development that could be used to enhance the quality of case study research. As Yin (2003:2) states: “Case studies can help us to inductively understand social life from the perspective of the actor as well as deductively test theories generated from this understanding. This gives us a unique opportunity to understand complex social phenomena from different angles”. Critics of the case study method believe that the study of a small number of cases can offer no grounds for establishing reliable findings that can be generalised. Others feel that, due to the intense exposure to the case being studied, leads to biased findings. Some dismiss case study research as useful only as an exploratory tool. Yet researchers continue to use the case study research method with success in carefully planned and crafted studies of real-life situations, issues and problems. Reports on case studies from many disciplines are widely available in the literature.

NOTES 1 See Schurink, 2009. 2 “Intellectual puzzles can and do take a variety of forms connected to the ontological and epistemological positions encapsulated in the research, and grounded within the specific context of their research problem. It is also the case that different theoretical and intellectual traditions in the social sciences are preoccupied with different kinds of intellectual puzzles, and consequently different kinds of social explanation” (Mason, 1996:15).

BIBLIOGRAPHY Baxter, P. and Jack, S. 2008. Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report. 13(4): 544–559. Bailey, K.D. 1978. Methods of Social Science Research. New York: The Free Press. Cohen, D.J. and Crabtree, B.F. 2008. Evaluative criteria for qualitative research in health care: Controversies and Recommendations. Annals of Family 6, Medicine available at http://www. annfammed.org/cgi/content/full/6/4/331. Creswell, J. 2007. Qualitative inquiry & research design: choosing among five approaches. Thousand Oaks: SAGE. David, M. (Ed.). 2006. Editor’s Introduction. Case study research: SAGE benchmarks in social research methods. Los Angeles: SAGE. Denzin, N.K. 1978. The Research Act: A Theoretical Introduction to Sociological Methods. New York: McGraw-Hill. Denzin, N.K. and Lincoln, Y.S. 2005. Introduction: The discipline and practice of qualitative research. In Denzin, N.K. and Lincoln, Y.S. (Eds), The SAGE Handbook of Qualitative Research (3rd ed.) London: Sage. E. Schurink & C.J. Auriacombe

453

Denzin, N.K. and Lincoln, Y.S. 2000. Handbook of qualitative research. (2nd Ed.), London: SAGE Publications. Eisenhardt, K.M. 1989. Building theories from case study research. Academy of Management Review, 14(4):532–50. Etherington, K. 2006. Becoming a reflexive researcher: using ourselves in research. London: Jessica Kingsley. Flick, U. (Eds.) 2007. Designing qualitative research. The SAGE qualitative Research Kit. Thousand Oaks, CA: Sage. Flyvbjerg, B. 2006. Five Misunderstandings about Case Study Research. Qualitative Inquiry, 12(2):224–228. Geertz, C. 1973. The interpretation of cultures. Basic Books, New York. Giogi, A. 1988. Validity and reliability from a phenomenological perspective. In Baer, Moss, L.P., Rappard, H. and Stam, H. (Eds.). Recent trends in theoretical psychology. New York: SpringerVerlag. Glaser, B.G. and Strauss, A.L. 1967. The discovery of grounded theory. Aldine: New York. Glaser, B.G. and Strauss, A.L. 1967. The discovery of Grounded Theory. Chicago: Aldine. Glaser, B.G. 1978. Theoretical Sensitivity: Advances in the methodology of Grounded Theory. Mill Valley, CA: Sociology Press Graham, B. 2000. Case study research methods. London: Continuum. Gummesson, E. 2007. Case study research and network theory: Birds of a feather. Qualitative Research in Organizations and Management, 2(3):226–248. Hammersley, M. and Atkinson, P. 1995. Ethnography: Principles in practice. London: Routledge. Hammersley, M. 2010. A historical and comparative note on the relationship between analytic induction and grounded theorising. Qualitative social research, 11( 2). Holloway, I. and Wheeler, S. 2002. Qualitative research in nursing. (2nd Ed.) Oxford, UK: Blackwell. Kavale, S. 1996. Interviews: An introduction to qualitative research interviewing. Thousand Oaks: Sage. Lincoln, Y.S. and Guba, E. 1985. Naturalistic Enquiry. Beverly Hill CA: Sage. Lindesmith, Alfred R. 1947. Opiate addiction. Evanston ILL: Principia Press. Marshall, C. and Rossman, G.B. 1989. Designing qualitative research. Newbury Park CA: SAGE. Mason, J. 1996. Qualitative researching. Thousand Oaks: SAGE. Mauthner, N. S. and Doucet, A. 2003. Reflexive accounts and accounts of reflexivity in qualitative data analysis. Sociology, 37(3):413–431. Maxwell, J.A. 2005. Qualitative research design: an interactive approach. (2nd ed). Thousand Oaks, CA: SAGE Merriam, S.B. 1998. Qualitative research and case study applications in education. San Francisco: Jossey-Bass.

454

Journal of Public Administration • Vol 45 no 3 • September 2010

Miles, M.B. and Huberman, A.M. 1994. Qualitative data analysis: An expanded sourcebook (2nd Ed.). Thousand Oaks: SAGE. Mouton, J. Marais, H.C. 1996. Basic concepts. Pretoria: HSRC. Paré, G. 2002. Enhancing the rigor of qualitative research: Application of a case methodology to build theories of IT implementation. The Qualitative Report, 7(4): available at http://www.nova. edu/ssss/QR/QR7-4/pare.html. Pandit, N.R. 1996. The Creation of Theory: A Recent Application of the Grounded Theory Method. The Qualitative Report, 2(4), available at http://www.nova.edu/ssss/QR/QR2-4/pandit.html. Remenyi, Dan, Money, Arthur, Price, David, and Bannister, Frank 2002. The creation of knowledge through case study research. Irish Journal of Management, 23(2):1–17. Schatzman, L. and Strauss, A.L. 1973. Field research; strategies for a natural sociology. New York: Prentice Hall. Schram, T. 2006. Conceptualizing and proposing qualitative research. Upper Saddle River: Pearson Education, Inc. Schurink. 2009. The internal audit as a tool to enhance the quality of qualitative research. Journal of Public Administration, 44(4.2):788–802. Schwandt, T.A. 1996. Farewell to criteriology. Qualitative inquiry, 2(1):58–72. Schwandt, T.A. 2007. The Sage dictionary of qualitative inquiry. Thousand Oaks: Sage. Seale, C. 1999. Quality in Qualitative Research. Qualitative Inquiry, 5(4):465–478. Shahaliizadeh, M., Amirjamshidi, G. and Shahalizadeh, S. 2009. Benchmarking of thesis research: a case study. Benchmarking: An International Journal, 16(1):103–124. Shank, G. 2006.Qualitative research: a personal approach. Ohio: Pearson. Simons, H. 2009. Case Study Research in Practice. London: SAGE: 164–167. Stake, R. 1995. The art of case study research. London: SAGE. Stiles, W.B. 1993. Quality control in qualitative research. Clinical Psychology Review, 13:593–618. Strauss, A. and Corbin, J. 1990. Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park: SAGE. Yin, R.K. 1994. Case Study Research. Design and Methods Thousand Oaks: SAGE. Yin, R.K. 1999. Enhancing the quality of case studies in health services research. Health Services Research, 34(5): Part II:1209–1224. Yin, Robert K. 2003. Case study research, design and methods (3rd Ed., Vol. 5). Thousand Oaks: Sage.

E. Schurink & C.J. Auriacombe

455