Description
Technology (from Greek τέχνη, techne, "art, skill, cunning of hand"; and -λογία, -logia[1]) is the making, modification, usage, and knowledge of tools, machines, techniques, crafts, systems, and methods of organization, in order to solve a problem, improve a preexisting solution to a problem, achieve a goal, handle an applied input/output relation or perform a specific function.
Research Reports in Reflecting on the Role of Literature in Qualitative Public Administration
ABSTRACT When undertaking qualitative research, public administration scholars must walk a thin line between being theoretically sensitive and imposing preconceived ideas on their work. This article identifies opportunities and pitfalls in using literature in qualitative public administration research. Whereas the opportunities are already well known within the discipline, the pitfalls remain underexposed. We identify potential pitfalls by using insights from the grounded theory approach. To illustrate how opportunities can be optimally exploited, and pitfalls avoided, we provide examples of high quality public administration research. Finally, we derive recommendations for public administration scholars when using literature in their qualitative research. These recommendations can help improve qualitative methods in the public administration discipline.
2
1
PUBLIC ADMINISTRATION RESEARCH AND THE ROLE OF LITERATURE
Textbooks on methods for public administration research contain many arguments in favor of using literature in the research process (e.g. McNabb, 2002:67-68; Bearfield & Eller, 2007:61-63; Justice, 2007:83-84). A literature review of publications that deal with the content of a research field, that is non-methodological literature, such as the key concepts, existing theories, theoretical frameworks, and/or empirical results, serves many purposes. For example, it helps researchers formulate a framework for analyzing their topics and problems (Justice, 2007:83). Further, the literature "provides clarity on a given subject by revealing long standing conflicts and debates, reveals the interdisciplinary nature of research on a subject, and places the work in a historical context" (Bearfield & Eller, 2007:62). Using the literature can have many advantages. However, contrary to what many public administration scholars believe, the overall advantage of using literature in all phases of the qualitative research process is not as self-evident as they suppose because it also comes with potential disadvantages. For instance, an extensive literature review may form a barrier to developing fresh insights, as a researcher may impose "predetermined understanding and existing frameworks on the investigation" (Heath, 2006:519). Further, researchers may select cases that accord with the existing literature and avoid cases or data that may provide contradictory evidence (Heath, 2006:523). If this occurs, then existing theory will have structured the process of data gathering in an undesirable way (Becker, 1993). Hence, questions related to when and how to use literature during the research process are important methodological issues. The issue of when and how to use the literature in qualitative research is relevant to quite a number of social science disciplines. Nevertheless, we argue that it should be
3
of particular interest to public administration scholars for two reasons. Firstly, although there are potential disadvantages in using literature in all qualitative research, it is especially difficult to find arguments against its use in the public administration discipline, or guidelines on how to deal with the dangers of excessive use. Handbooks on methods for public administration rightly indentify the literature review as a crucial step in the design and conduct of research (e.g. McNabb, 2002:393), but they generally fail to recognize the importance of the methodological considerations related to timing and implementation surrounding it. As such, the pitfalls of using literature remain underexposed. Secondly, when and how to use the literature in qualitative research carries particular importance for the discipline of public administration since the quality of qualitative research in this field has been called into question. Brower, Abolafia, and Carr (2000:363), for example, state with regard to public administration as a discipline that "the field's use of qualitative methods reveals substantial weaknesses". They offer four general guidelines for improving qualitative methods in public administration research: "push description to explanation", "improve the quantity of data", "develop theoretical sensitivity", and "develop sensitivity for the backstage" (Brower et al., 2000:387-390). Under the third guideline, they explicitly recommend public administration scholars to "exercise caution to allow the regularities and anomalies in the data to suggest possible theories, rather than force data into theories in ways that obtain a premature closure of meaning" (idem, 389). We elaborate on this recommendation by focusing specifically on the use of the literature and by developing specific lessons for the qualitative public administration research process. The way one deals with the literature plays a crucial and guiding role in the iterative process between theory and data because it determines how existing theories are dealt with in the design and conduct of research. As such, considerations as to when and how to use the literature acquire great importance. A
4
more thorough reflection on the role and use of literature in research in the public administration field may consequently improve the reception and application of public administration research, the legitimacy and usefulness of which according to some remain questionable (e.g. Bogason & Brans, 2008:92). These considerations justify paying significant attention to the role of literature in public administration research. To determine guidelines for how to deal with this important issue, one has to look beyond the public administration discipline. Many of the more critical scholars come from the fields of health and nursing studies. They often use the grounded theory approach, which is very critical of the use of literature (Glaser & Strauss, 1967).i According to Glaser (1998:67), a strong dictum within grounded theory is "do not do a literature study in the area and related areas where the research is to be done". Morse (2002:295), for example, states that the "fear of invalidating one's own work by violating the principles of induction is prevalent in the writing of qualitative researchers. In fact, whether or not to go to the library before beginning fieldwork has become a researcher's dilemma". This somewhat extreme view is often disputed by grounded theorists but, at the very least, many scholars do not take the use of literature for granted (e.g. Becker, 1993; Cutcliffe, 2000; Morse, 2002; Heath & Cowley, 2004). We draw insights from the grounded theory approach as our starting point, and use these to identify the potential pitfalls when using literature, even when using qualitative approaches other than grounded theory. The potential opportunities are derived from both grounded theory and from the public administration literature. We explicitly choose to speak in terms of opportunities and pitfalls, rather than, for example, in terms of advantages and disadvantages. This is because we believe that a good researcher is able to minimize the disadvantages of a particular method and to fully exploit its advantages. From this overview of pitfalls and opportunities, we derive
5
recommendations for public administration scholars when using literature in qualitative research. Our focal question is:
What are the opportunities and pitfalls when using literature, and what recommendations for public administration scholars undertaking qualitative research can be derived from these?
The outline of this article is as follows. First, we introduce grounded theory and consider its perspective on the use of literature. Second, we identify a number of opportunities and pitfalls. We illustrate these by providing concrete examples taken from public administration research that deal with the literature effectively. Third, we formulate recommendations for public administration scholars to follow when using literature in qualitative research. 2 THE GROUNDED THEORY APPROACH
In this section, we outline the general characteristics of 'the' grounded theory approach. A difficulty is that, since the methodology was first developed in the 1960s, two very different approaches have evolved, namely those of Glaser and of Strauss. The methodological differences between these two approaches are substantial. Some academic scholars even argue that two fundamentally different research strategies have evolved (Glaser, 1992:2; Wilson & Hutchinson, 1996; Cutcliffe, 2000:1481 ff.). However, we do not share this view. What has evolved, we would argue, are two methodological approaches that belong to the same research strategy. The two approaches share essential defining characteristics that make them grounded theory approaches (see also Benoliel, 1996; Melia, 1996; Cutcliffe, 2000; Walker & Myrick; 2006:549-550; Bryant &
6
Charmaz, 2007). Before going into the detail of the grounded theory approach, we provide the background to its development and its relationship with other approaches.
2.1
Background of the grounded theory approach
The grounded theory approach was developed in the 1960s by Barney Glaser and Anselm Strauss (Glaser & Strauss, 1967; see also Boychuk Duchscher & Morgan, 2004:606). In those years, qualitative sociological research was losing ground to more quantitative methods. This shift was closely related to the increasing dominance of the positivistic paradigm. Charmaz (2006:4) notes that "only narrowly scientific - that is, quantitative - ways of knowing held validity for positivists; they rejected other possible ways of knowing, such as through interpreting meanings or intuitive realizations". In their book The discovery of grounded theory, Glaser and Strauss (1967) questioned those epistemological assumptions. They made an innovative statement, contesting those notions and at the same time offering systematic qualitative research strategies (Charmaz, 2006). In essence, they argued that research flowing from the positivistic stance had led to the development of abstract theories that no longer did justice to empirical realities. As an alternative, they proposed a method for systematic qualitative analysis that would generate theories that were 'grounded' in empirical realities. The impact of their book on qualitative research has been substantial. It was firstly influential in nursing and health studies, most prominently those undertaken at the University of California. Over the years, scholars in many other fields, such as public management and the sociology of professions, have embraced this research approach (Charmaz, 2006). For instance, today, professionals involved in student affairs use
7
grounded theory to increase educators' understanding of the complex interactions between students and college environments (Bowen, 2008). Examining its relationship with other research methods, grounded theory can be positioned among other inductive, qualitative approaches (Strauss & Corbin, 1990). In inductive analyses in general, patterns, themes, and categories of analysis come from the data rather than from predetermined theory. Theory is derived from the data, rather than being imposed on the data prior to data collection or analysis (Patton, 1980). That is, a grounded theory is derived inductively, through the systematic collection and analysis of data using a continuous comparative approach. Related approaches - which do differ from grounded theory in a number of aspects - include phenomenology and ethnography. At the other end of the spectrum, there are distinctly different approaches such as deductive, qualitative or quantitative, analyses. Deductive approaches start from a theory and subsequently test this theory using empirical analyses. Grounded theory explicitly runs counter to these more deductive approaches, especially to those that use quantitative methods.
2.2
Characteristics of the grounded theory approach
Having provided a short background to grounded theory, we will now describe its main characteristics. The first important characteristic of the grounded theory approach has already been mentioned: grounded theory is inductive (Eaves, 2001:655; Heath & Cowley, 2004:142). The second key characteristic is that the grounded theory approach has a specific way of developing theories, namely through "purposeful systematic generation from the data of social research" (Glaser & Strauss, 1967:28). Theories are constructed from the data themselves (Eaves, 2001:655-656). The supposed advantage
8
of this is that the theories that are being developed, almost by definition, are grounded in the data and therefore do justice to the social reality. The third general characteristic of grounded theory is that it aims to develop explanatory theories which, as such, are very different from descriptive or purely conceptual theories. Alongside these three general characteristics, two shared grounded theory characteristics can be identified that relate to the methodology. The first methodological characteristic shared by both grounded theory approaches is that of constant comparison. During a grounded theory study, data are continually compared with: 1) other primary data, 2) evolving original data, 3) evolving concepts, and 4) evolving theories (see Boychuk Duchscher & Morgan, 2004:607; see also Glaser & Strauss, 1967). In making these comparisons, grounded concepts and the relationships between them become visible. The second methodological characteristic shared by both grounded theory approaches is the process of 'open coding' of data. By assigning codes that have not been fixed in advance to data during the analysis, and by a continuous comparison between codes, grounded theorists develop concepts that form the basis of evolving theories. As already noted, two distinct grounded theory approaches have evolved since the 1990s. The differences between the Glaserian and the Straussian approaches spring from an intense debate between the two on what grounded theory should look like. In 1990, Strauss, in cooperation with Juliet Corbin, published Basics of Qualitative Research. This book is at the heart of the debate between Glaser and Strauss (Eaves, 2001:656). Strauss introduced a new coding method: "a coding paradigm involving conditions, context action/interactional strategies and consequences" (Strauss & Corbin, 1990:67). Even though both approaches adhere to the principle of 'open coding', the differences between the respective coding methods of Glaser and of Strauss are substantial (see Strauss & Corbin, 1990; Kendall, 1999; Eaves, 2001; Boychuk
9
Duchscher & Morgan, 2004; Glaser & Holton, 2004; Heath & Cowley, 2004; Walker &
Myrick, 2006). Rather than letting theory emerge from the data, Strauss 'forces' the data and the theory into a methodological framework which, it is argued by critics of this approach, violates the principle of induction (see also Boychik Duchscher & Morgan, 2004). According to Glaser, this has led Strauss away from the basics of grounded theory and toward a method that can no longer be considered as grounded theory (Glaser, 1992:2). 2.3 The use of existing literature in grounded theory
One aspect of the debate between Glaser and Strauss concerns the question as to when and how literature should be used in the research process. In this article, this is an important topic as it is closely linked to our aim, which is to see what lessons can be learnt for public administration research from the debate among grounded theorists on the use of literature in qualitative research. First and foremost, it is well worth noting that the role of literature is relatively minor in both Glaser's and Strauss' approaches (Baker et al., 2002). Essentially, the ideas underpinning grounded theory are very critical of its use, and both approaches remain loyal to this sentiment. Since grounded theory is designed to inductively derive theories from the data, existing literature should be viewed with suspicion at the very least. Grounded theorists generally fear that they will violate the principle of induction as a consequence of using literature either too early or in the wrong way (e.g. Glaser, 1998:68-69). This is not an imagined fear; there is a real danger that a researcher will violate one of grounded theory's main principles through using literature: theories should be derived from the data, and from nothing else. However, most grounded theorists do recognize the usefulness of literature in at least some phases of a research project, and
10
do try to incorporate its use in their methods, albeit in very different ways (Heath & Cowley, 2004:143). In essence, Strauss 'allows' the existing literature to function as a framework for analysis early in the research process; whereas Glaser only uses the literature later, and then only in the role of providing comparative data. Here, for two reasons, we will not discuss in detail the differences in use of the literatureii, but instead focus on the arguments that are being advanced in the debate. Our first reason is that the place and function of a literature review in Glaser's and Strauss' approaches are intensely debatediii and not unambiguous (Charmaz, 2006:165). Second, we are not so much interested in the methodological aspects of grounded theory, but more in the arguments that can play a role in the ongoing debate about the use of literature in grounded theory. Clearly, this 'methodological battle' is far from over. Nevertheless, the methodological literature on grounded theory provides a valuable collection of arguments for and against the use of literature in different phases of grounded theory studies. Our aim is to see what lessons can be learnt from this debate that could be applied to other qualitative research strategies that are used in the field of public administration. We believe that the arguments advanced could function as a valuable mirror for public administration scholars. Both Glaser and Strauss, and their respective followers, provide insights that, while they primarily apply to grounded theory methodology, can be transposed in such a way that they shed light on the opportunities and pitfalls of using literature in public administration research in general. Using Glaser's and Strauss' arguments as a guide, this is what makes grounded theory such a useful perspective in more generally reflecting on the role of literature in qualitative research. What is innovative about our approach is that it explicitly combines both grounded theory approaches, whereas previous reflections on the role of literature have, either explicitly or implicitly, tended to take either a Glaserian or a Straussian perspective and, furthermore, were usually limited to a grounded theory perspective. We believe that the
11
two perspectives can be combined, and that there are also lessons to be drawn that are relevant for public administration scholars. 3 OPPORTUNITIES AND PITFALLS IN USING LITERATURE DURING EACH SUCCESSIVE RESEARCH PHASE
In this section, we begin to identify the opportunities and pitfalls in using literature. This discussion is not so much focused on the question as to whether researchers should read and review the literature: without doubt, researchers have to relate their work to the existing literature in their field. The opportunities and pitfalls are more concerned with when to use literature, and if and how to use it in analysis. Glaser and Strauss both argued for delaying, rather than abandoning, the literature review (Charmaz, 1990:1163). Grounded theory can warn us of the opportunities and pitfalls in reviewing literature early in the research project, i.e. before the initial analysis has been completed (see also Glaser, 1998:67-80). Given that the opportunities and pitfalls are (partly) related to when one uses the literature, we have categorized them according to the research phase in which they are most likely to occur. We distinguish three main phases, which we will discuss separately below: research design (3.2), data collection (3.3), and data analysis (3.4).iv 3.1 Selecting illustrative public administration examples
The opportunities and pitfalls we have identified can remain quite abstract. Throughout this section we therefore discuss research drawn from the public administration discipline that effectively used - or deliberately did not use - a literature review in the various phases of the research process. For each opportunity, and for each pitfall, we give an example of good quality public administration research and we briefly explain
12
how the researcher minimized the effects of the applicable pitfall, or exploited the opportunity. Note that we do not consider these examples to be best practices. Rather, the selection includes those studies which we believe to be the most instructive for the reason that the authors provide an elaborate account of the considerations behind the way in which they used literature. We hope doctoral students and established scholars find inspiration in these examples, which could function as illustrations of how to handle literature reviews in qualitative research. We used four methods to gather examples. First, we searched scientific databases. As different databases scan different sets of journals, we used a number of databases: ABI/Inform Global, JSTOR, Sage Journals online, and ISI Web of Knowledge. Topics that were used in searching these databases included [public administration], [political science], [policy studies], [qualitative], [inductive], [grounded theory], [deductive], [literature] and [literature review], as well as a number of synonyms. Second, we analyzed the abstracts of articles published between 2005 and June 2009 in five leading public administration journals (Administration & Society, Journal of Public Administration Research & Theory, Public Administration, Public Administration Review, and Public Management Review). Third, we used the descendancy approach (Bowen & Bowen, 1999:74), which amounts to locating all studies cited in key publications. We used Brower et al. (2000) as our key publication. Fourth, we asked fifteen Dutch public administration scholars, five of whom can be considered to be methodological experts, to provide good quality research examples for each of the identified opportunities and pitfalls. Four of them provided suitable examples. We used four main selection criteria. First, we only selected research from the field of public administration. Second, we only chose studies that took a qualitative approach. Third, only publications in English were considered. Fourth, the research
13
report had to pay explicit attention to considerations regarding the use of literature. The selections made of this basis include those studies we believe to be the most instructive. Note that not all authors were claiming to be using grounded theory, with the notable exceptions of Agranoff (2007), DeHart-Davis (2008), and De Kort and Van der Pijl (2009). As such, we are also drawing on other research strategies, since these can also provide valuable insights into the use of literature. 3.2 Phase 1 - Research design
In the initial research design phase, the researcher formulates a research question, selects a methodological approach, and then operationalizes the relevant variables (McNabb, 2002). In this phase, we have identified three opportunities and one pitfall related to literature use. A first major, and well known, opportunity is that reading literature helps researchers identify 'knowledge gaps' (Hutchinson, 1993). Subsequently, it becomes easier to connect a research project to other research, to show the theoretical relevance of a study, and to integrate the evolved theory with other theories. The added value of a study, that is its contribution to the body of knowledge, hence increases. Grasping this opportunity can be illustrated using Ackroyd, Kirkpatrick, and Walker's study of public management reform in three services in the UK: healthcare, housing, and social services (2007). In their article, they note that "In the literature there are now many overviews of this process [of public management reform], but few systematically compare the effects of restructuring. With some exceptions (see Boyne et al., 2003)v accounts tend either to analyze public services as separate cases or emphasize general trends in them all. Hence, what is often lacking is a 'comparative analysis across different domains' (Ferlie et al., 2003:S3)". Here, Ackroyd and his coauthors use the existing literature to identify a knowledge gap, arguing that comparative
14
research in their field is lacking. Their study then starts to compare public management reform in three UK services. Through this, they add theoretical relevance to their study. If they had not used existing literature, it would have been more difficult for them to connect their project to other research, and it could be the case that they would have researched something which was already well-known within the discipline. Secondly - and partly related to the first opportunity - the use of literature can help to explain to the reader why a particular question is important. It helps in answering the 'so what' question. Although a question may sound interesting, or provoke curiosity, to properly establish its importance, research must be placed in context. A wellconstructed literature review can provide a panoramic view of the research question and help the reader understand why a particular question matters (Bearfield & Eller, 2007:62). This can be illustrated using the scholarly work of Esther Versluis Enforcement Matters: Enforcement and compliance of European Directives in four Member States (2003). This is a comparative analysis, involving four member states of the European Union, of the transposition and enforcement of, and compliance with, two directives regulating dangerous chemical substances. With reference to existing literature, she shows that a comparative study of compliance and enforcement is important. She for example states that "Strict regulation can lose its meaning when enforcement agencies keep an eye closed when monitoring companies or when they make no use of sanctions. Obedience differs between countries - 'In some countries, rules are there to be followed, in others it seems if they merely exist to be violated.' (Van Waarden, 1998:2) - and insufficient monitoring of the enforcement process by the Commission probably leads to diverging enforcement practices within the Member States." (Versluis, 2003:10). She then continues to formulate her research questions on this topic of possible diversity in compliance and enforcement. Hence, she is placing the research into context and
15
highlighting why the particular research questions are important. If she had not referred to existing literature to highlight why a research question was relevant, it would have been far more difficult to answer the 'so-what' question: why does this research matter, and how does it add to existing knowledge? Hence, the way she used existing literature provided legitimacy for conducting large-scale research on comparing compliance and enforcement in a number of EU member states. Finally, using the existing literature can help focus a research project (McNabb, 2002:67). Literature can significantly narrow the focus of a research project by providing tentative questions or enhancing conceptual clarity. Cutcliffe, (2000:1480), using his own doctoral study on hope as an example, notes that "Many proposed research questions require conceptual clarity. [A literature] review may help provide a sense of the key elements of hope that are implicit in the literature, it may help provide some conceptual clarity of the nature of hope and the nature and practice of bereavement counseling and this examination of the relevant literature would help the researcher to differentiate hope from similar and related concepts." In their recent JPART article, Foldy and Buckley (2010) examine team learning among street-level bureaucrats. State-of-the-art literature from the fields of social psychology, organizational behavior, and public administration was arguing in favor of focusing on the group level, rather than for instance on individual characteristics, if one is to effectively examine team learning. Therefore, the authors of this article decided to examine teams, documenting whether they discarded old routines and learned new ones. Also in other phases of their research, Foldy and Buckley often used literature to narrow their focus. They, for example, constructed hypotheses based on scholarly work: "Drawing on previous research on both team effectiveness and learning, we developed hypotheses on the team characteristics that could have an impact on team learning." (ibid:25). It is worthwhile considering what could have happened if they had not used the
16
existing literature to provide focus. If they had not focused on the group level, this could have resulted in a less clear-cut article, maybe one looking at individual characteristics, organization characteristics and team characteristics. This would have provided few insightful additions to the scholarly literature and for practitioners. However, using literature in this phase of research also has a substantial pitfall. According to Becker (1993) and Heath and Cowley (2004), a researcher runs the risk of overlooking social phenomena, or relevant aspects, by focusing exclusively on those issues that seem relevant according to the literature. Referring to the work of Glaser (1998:67-73), Heath and Cowley (2004:143) note that researchers should be aware of "near misses in discovering new theory, [which] is a process whereby as the theory begins to emerge, literature of close relevance is recognized or read and its powerful impact bends the emerging theory from its true path." This can also be illustrated using an example from the public administration discipline. According to numerous EU studies, nationality forms a basis for networks in the European Commission. However, authors do not question this relationship, and therefore do not systematically assess this assumption. Suvarierol's book Beyond the myth of nationality (2007) goes further and challenges the current EU literature by systematically exploring the role of nationality and networks in the European Commission. In one important section, entitled 'Debunking the myth of nationality', Suvarierol concludes that "The results can be summarized as follows. Even though many officials recognize the mechanisms that underlie homophily resulting from cultural similarities, the networks of officials are not culturally homophilic." (ibid:104). Essentially, she is stating that the networks of these officials are not as nationally or culturally oriented as is often assumed. In her book, Suvarierol focuses on other social phenomena which are generally not considered relevant in the literature, thereby avoiding the pit that others had been steered into. If, as other scholars, she had
17
accepted this nationality myth at face value, she would most likely have missed important ways in which EU officials work in Brussels. She would have simply assumed that national networks play a major role, and maybe would have interpreted her data using this assumption. Consequently, examining this often accepted phenomenon enabled Suvarierol to debunk some common myths about the way officials work in the European Union. 3.3 Phase 2 - Data collection
In the data collection phase, data are gathered in order to satisfy the research objectives and answer the research question (McNabb, 2002:56). An opportunity in using literature in the data collection phase is that it can provide an additional data source, alongside, for example, interviews and observations. This corresponds with Glaser's notion that "all is data" (Glaser, 2001:145). In this way, the researcher can gain a greater understanding of the relevant phenomena. For the purposes of this article, it is unfortunate that the number of public administration studies that include existing literature as another data source is low. One of the few exceptions from SSCI ranked journals that we found is a 2009 article by De Korte and Van der Pijl. The authors included existing articles in both the data collection and the data analysis phases of their study on organizational change within the setting of a governmental bureaucracy. Regrettably, the role of these articles is not specified in great detail, which makes it difficult to assess the effect of including existing literature as another data source in the study. Our interpretation is that the additional 'data' enriched their analysis by providing codes that could be compared with other codes that emerged from other data. On the other hand, using literature in this phase has a major pitfall. Researchers may go on to select cases (or data within cases) for their research that accord with the
18
existing literature and avoid cases or data that may provide contradictory evidence (Heath, 2006:523). If this occurs, existing theory will have structured the process of data gathering in an undesirable way (Becker, 1993). The end result of such a data collection process may be a biased picture - one that does not correspond with reality. In his study of so-called 'public management networks', Agranoff (2007) used a grounded theory approach because he "had to dig deeper into the data, looking for a form of meaning that might lie outside the previous literature" (ibid:36). In his view, he needed to go "beyond the trendy assertions" (ibid:35) of existing, well-established network theories that held that networks functioned as decision-makers. Agranoff argued that although existing theories could function as value frameworks for analysis, they lacked a perspective on those networks that do not make decisions. In an alternative approach, Agranoff set out to uncover network management as practiced by its practitioners. If, in the data collection phase of the research, he had only concentrated on data that were relevant according to the existing literature he would most likely have missed important roles of public management networks other than decision-making, because the existing literature provided little room for these. Subsequently, grounded theory enabled Agranoff to generate the contextualized knowledge that existing network theories were lacking in his view (see also Bogason & Brans 2008:92-93). 3.4 Phase 3 - Data analysis
Once the data are available, the researcher has to order the data and determine their meaning. This interpretation is necessary because it enables the findings to be related to the original research question and research objectives (McNabb, 2002:57-58). In this data analysis phase, one opportunity gained from using literature is that researchers can enhance their insights into the object under study (Cutcliffe, 2000:14801481; Strauss & Corbin, 1990). For example, an insight may be enhanced through the
19
literature increasing a researcher's sensitivity to latent relationships within the data: that is, it strengthens theoretical sensitivity (Strauss, 1987; Heath & Cowley, 2004:143; see also Brower et al., 2000:389). In this way, prior comprehension plays a positive role in analyzing the new data (Heath, 2006). This helps the researcher to achieve the main goal of qualitative research: "Verstehen" (understanding). In 1971, Allison published his famous analysis of the Cuban missile crisis. Instead of only using rational expectancy theories, Allison used two additional models to explain how the US and the Soviet Union made decisions in this Cold War event. The organizational behavior model and the government politics model enabled him to enhance his insights, because each model focused on particular relationships and concepts that were more-or-less absent in the other models. The theoretical models that Allison derived from the existing literature functioned as supplementary frames of reference (Allison & Zelikow, 1999:5), providing him with additional interpretations of decision-making processes. This approach provided a much more all-encompassing view on how decisions were made. This research strategy enabled Allison to draw on existing literature in order to understand what had happened. If he had not used the results of his literature review in the analysis, he would not have benefited to the same extent from such well established framework models, including their core theoretical concepts that increased his sensitivity to latent relationships within the data. Conversely, using literature in this phase also has a pitfall. It can form a 'barrier of insight' (Heath, 2006:520; Glaser, 1978), with the result that a researcher interprets data through a framework provided by the literature. Consequently, a researcher may impose "predetermined understanding and existing frameworks on the investigation" (Heath, 2006:519, see also Glaser, 1978; Cutcliffe, 2000:1481; Stern, 1994). In other words, researchers import preconceived ideas - such as existing theoretical structures and impose them on their own work (Charmaz, 2006:165; Charmaz, 1990:1163; cf.
20
Glaser, 1998:67 ff.). This can suppress fresh insights that might otherwise have emerged from the data. DeHart-Davis (2008:363-364), for example, feared that the extensive body of literature on the nature and consequences of ineffective rules, or 'red tape', would form a barrier of insight in her study on effective organizational rules, dubbed 'green tape'. The red tape literature, in her view, had a strong presumption that rules inhibit creativity and flexibility. Further, it is said to suffer from a disproportionate focus on middle-level and senior managers. A study of 'green tape', DeHart-Davis argues, consequently runs the risk of anchoring respondents' perceptions of rule quality in pre-existing theory. To avoid this danger, she avoids using existing theories as a framework for her analysis, and instead identifies patterns in the data through a grounded theory approach. Had she used the existing literature as a starting point for analysis of her interview data, it would have been difficult to detect the attributes of green tape because the literature provides little room for such rules. It was for that reason that DeHart-Davis employed a grounded theory research strategy which allowed her to indentify and categorize the attributes of green tape without imposing preconceived ideas of what are 'good' or 'bad' rules on her work. For another good example of avoiding barriers of insight, see Durant's (2007) study on large-scale organizational change in public organizations. He conducted extensive, inductive archival research in an "understudied arena of public policy" (idem, 410) and this strategy enabled him to avoid imposing preconceived ideas on his work. 3.5 Summarizing opportunities and pitfalls
Based on the discussion above, we have constructed Table 1 which summarizes the opportunities and pitfalls in each research phase, and notes the related examples drawn from public administration literature discussed above.
21
Opportunities/pitfalls in each research phase
Public administration literature which used the opportunity optimally or effectively to avoid the pitfall
1. Research Design a. Opportunity: identifying knowledge gaps Ackroyd, Kirkpatrick, & Walker (2007) b. Opportunity: explaining why a particular research question is relevant ('answering the 'so what' question) c. Opportunity: focusing the research project Foldy & Buckley (2010) Suvarierol (2007) Versluis (2003)
d. Pitfall: Overlooking social phenomena
2. Data collection a. Opportunity: using literature as an additional data source b. Pitfall: selecting only data that are relevant according to the literature Agranoff (2007) De Korte & Van der Pijl (2009)
3. Data analysis a. Opportunity: enhancing insight by increasing sensitivity to latent relationships b. Pitfall: forming a barrier to fresh insights DeHart-Davis (2008), Durant (2007) Allison (1971)
22
4
RECOMMENDATIONS FOR PUBLIC ADMINISTRATION RESEARCH
We have detailed a number of opportunities and pitfalls related to using literature in new research. These have been established by drawing on literature on both grounded theory and public administration research. Further, we have provided examples from existing public administration research that demonstrated each of the opportunities and pitfalls. The main point being emphasized is that a benefit is not inevitable from using literature: there are both opportunities and pitfalls connected to using it. On this basis, we argue that public administration scholars should take a more critical stance toward the use of literature reviews in the various phases of the qualitative research process, and that grounded theory can provide inspiration in this direction. This conclusion has a number of consequences. First, we distinguish two general recommendations and we follow these by more-specific recommendations for the various stages of the research process.
General recommendations: The question as to whether to use literature or not deserves consideration in each phase of the qualitative research process. The use of literature in all research phases is not by definition a defining characteristic of sound research.
Recommendations for research design: In the research design phase, public administration scholars should indicate and justify where literature will be used in each of the remaining research phases, in
23
what role, and to what extent. Ideally, this should primarily depend on the research questions and the research goals. A thorough research design should further indicate how the potential pitfalls of using literature will be avoided and how the opportunities will be utilized to their full potential.
Recommendations for data collection and analysis: During data collection and analysis, the role of literature and its influence on the development of the research, and its results, should be continuously reviewed.
Recommendations for research reporting: Research reports should be explicit as to the phases in which literature has been used, and to what extent and how the opportunities and pitfalls have been addressed, and what limitations follow from this. Academic publications should similarly be explicit about how literature has been used during the research process, as well as about the limitations this imposes on the results.
To sum up, we have argued that explicitly questioning the role of literature is an essential part of the qualitative research process. If public administration scholars were to do this systematically, we believe that it would substantially enhance the quality of this research field. In our view, this applies to the diverse set of qualitative research approaches adopted in the public administration field and, as such, moves beyond grounded theory. It could strengthen the theory building capacity of the various research approaches, such as interpretivism, empiricism, rationalism, and postpositivism, used in the public
24
administration discipline as discussed by Riccucci (2010:65-96). She argues that all these forms of research could potentially contribute to theory building in public administration. We believe that these various research approaches can learn - to a greater or lesser extent - from recommendations derived in this article. This is because we do not generate specific do's and don'ts (which could be incompatible with some approaches) but, instead, encourage a more critical stance and awareness of the potential pitfalls in using literature in general. As such, we promote a continuous review of its use throughout the various research stages, and remaining aware and explicit about the choices that follow from that review. We end this article with a final, more practical, concern. One might worry that there may well be obstacles to pursuing these recommendations in terms of journal articles and research funding because, throughout the public administration discipline, literature reviews are usually regarded as an essential part of all research projects (see also Fink, 2005:6). This is an honest observation and, more generally, Schreiber (2001:57) argues that:
"The current expectations of academic research and funding agencies suggest that plunging into field research without delving into the relevant literature would be folly. [?] The researcher who is seeking funding must demonstrate an understanding of the 'state of the science' regarding the phenomenon of study in order for agency evaluators to be confident that their money would be wellspent."
These practical concerns can still be aligned with our recommendations as our message is not that literature reviews should not to be included in research designs or research reports. It is rather that the researcher should include a thorough explanation and
25
justification for the use of literature in the various phases of the research project. This should improve rather than diminish the quality of public administration research since it provides an underpinning to what should be a critical methodological consideration. Nevertheless, what will be required is a willingness by reviewers to fund and publish research that delays elements of the literature review until an appropriate research phase, recognizing that this may not necessarily be the design phase, provided that the researcher provides a sound methodological justification for the timing. This fits with a trend in which a growing number of public administration scholars employ interactive research strategies. According to Bogason and Brans (2008:92-93), this will help overcome the alleged cool reception and limited application of public administration theory in practice. Such developments should make funding, as well as publication, of public administration research easier.
26
REFERENCES
Ackroyd, S., Kirkpatrick, I. & Walker, R. M. (2007). Public management reform in the UK and its consequences for professional organization: A comparative analysis. Public administration, 85 (1), pp. 9-26.
Agranoff, R. (2007). Managing within networks: adding value to public organizations, Washington: Georgetown University Press.
Allison, G. & Zelikow, P. (1999). Essence of decision: explaining the Cuban missile crisis, Nw York: Longman
Baker, C., Wuest, J. & Stern, P.N. (2002). Method slurring: the grounded theory phenomenology example, Journal of advanced nursing, 17 (11), pp. 1355-1360.
Bearfield, D.A. & Eller, W.S. (2007). Writing a literature review: the art of scientific literature, in: Yang, K. & Miller, G.J. (eds.). Handbook of research methods for public administration, New York: CRC Press, Taylor & Francis, pp. 61-72.
Becker, P.H. (1993). Common pitfalls in published grounded theory research, Qualitative health research 3 (2), pp. 254-260.
Benoliel, J.Q. (1996). Grounded theory and nursing knowledge, Qualitative health research 6 (3), pp. 406-428.
27
Bogason, P. & Brans, M. (2008). Making public administration teaching and theory relevant, European Political Science, 7, p. 84-97
Boychuk Duchscher, J.E. & Morgan, D. (2004). Grounded theory: reflections on the emergence vs. forcing debate, Journal of advanced nursing 48 (6), pp. 605-612.
Bowen, G.A. (2008). Grounded theory and sensitizing concepts. International Journal of Qualitative Methods, 5(3), pp. 12-23.
Bowen, W.M. & Bowen, C.C. (1999). Typologies, indexing, content analysis, metaanalysis and scaling as measurement techniques, in: G. Miller & M. Whicker (eds.) Handbook of quantitative methods in public administration, New York, NY: Marcel Dekker, pp. 51-86.
Brower, R.S. Abolafia, M.Y. & Carr, J.B. (2000). On improving qualitative methods in public administration research, Administration & Society 32 (4), pp. 363-397.
Bryant, A. & Charmaz, K. (eds.) (2007). The SAGE handbook of grounded theory, London: SAGE.
Bulmer, M. (1979). Concepts in the analysis of qualitative data, Sociological Review 27 (4), pp. 651-677.
Charmaz, K. (1990). 'Discovering' chronic illness: using grounded theory, Social Science and Medicine 30 (11), pp. 1161-1172.
28
Charmaz, K. (2006). Constructing grounded theory: a practical guide through qualitative analysis, London: SAGE.
Cutcliffe, J.R. (2000). Methodological issues in grounded theory, Journal of advanced nursing, 31 (6), pp. 1476-1484.
DeHart-Davis, L. (2009). Green tape: a theory of effective organizational rules, Journal of Public Administration Research and Theory, 19 (2), pp. 361-384
Durant, R.F. (2007). Toxic politics, organizational change, and the 'greening' of the U.S. military: toward a polity-centered perspective, Administration & Society 39 (3), pp. 409446.
Eaves, Y.D. (2001). A synthesis technique for grounded theory data analysis, Journal of advanced nursing 35 (5), pp. 654-663.
Fink, A. (2005). Conducting research literature reviews: from the internet to paper, Thousand Oaks: SAGE
Foldy, E.G. & Buckley, T.R. (2010). Re-creating street-level practice: the role of routines, work groups, and team learning, Journal of Public Administration Research and Theory, 20 (1), pp. 23-52.
Glaser, B.G. (1978). Theoretical sensitivity: advances in the methodology of grounded theory, Mill Valley: Sociology Press.
29
Glaser, B.G. (1992). Emerging vs. forcing: basics of grounded theory analysis. Mill Valley: Sociology Press.
Glaser, B.G. (1998). Doing grounded theory: issues and discussions, Mill Valley: Sociology Press.
Glaser, B.G. (2001). The grounded theory perspective: conceptualization contrasted with description, Mill Valley: Sociology Press.
Glaser, B. G. & Holton, J. (2004). Remodeling grounded theory, Forum: Qualitative social research 5 (2), article 4, pp. 1-17.
Glaser, B.G. & Strauss, A.L. (1967). Discovery of grounded theory: strategies for qualitative research, Chicago: Aldine.
Heath, H. & Cowley, S. (2004). Developing a grounded theory approach: a comparison of Glaser and Strauss, International journal of nursing studies 41 (2), pp. 141-150.
Heath, H. (2006). Exploring the influences and use of the literature during a grounded theory study, Journal of research in nursing 11 (6), pp. 519-528.
Hutchinson, S.A. (1993). Grounded theory: the method, in: Munhall, P.L. & Boyd, C.O. (eds.). Nursing research: a qualitative perspective, New York: National League for Nursing, pp. 180-212.
30
Justice, J.B. (2007). Purpose and significance of research design, in: Yang, K. & Miller, G.J. (eds.). Handbook of research methods for public administration, New York: CRC Press, Taylor & Francis, pp. 75-92
Kendall, J. (1999). Axial coding and grounded theory, Western journal of nursing research 21 (6), pp. 743-757.
Korte, R. de, & Pijl, G. van der (2009). Characteristics of a central change programme within a governmental bureaucracy: a grounded theory study, Journal of Management & Governance 13 (1-2), pp. 5-40
Melia, K.M. (1996). Rediscovering Glaser, Qualitative Health Research 6 (3), pp. 368378.
McNabb, D.E. (2002). Research methods in public administration and nonprofit management: quantitative and qualitative approaches, Armonk: ME Sharpe.
Morse, J.M. (2002). Theory innocent or theory smart?, Qualitative health research 12 (3), pp. 295-296.
Morse, J.M. & Field, P.A. (1995). Qualitative research methods for health professionals, California: SAGE.
Patton, M.Q. (1980). Qualitative evaluation methods, Beverly Hills: SAGE.
31
Riccucci, N.M. (2010). Public administration: traditions of inquiry and philosophies of knowledge, Georgetown: Georgetown University Press.
Schreiber, R.S. (2001). The "how-to" of grounded theory: avoiding the pitfalls, in: Schreiber, R.S. & Stern, P.N. (eds.), Using grounded theory in nursing. New York: Springer.
Stern, P. N. (1994). Eroding grounded theory, in: J.M. Morse (ed.) Critical issues in qualitative research methods, Thousand Oaks: SAGE, pp. 212-223.
Strauss, A.L. (1987). Qualitative analysis for social scientists, New York: Cambridge University Press.
Strauss, A.L. & Corbin, J.M. (1998). Basics of qualitative research: techniques and procedures for developing grounded theory, London: SAGE.
Strauss, A.L. & Corbin, J.M. (1990). Grounded theory research: procedures, canons and evaluative criteria, Qualitative sociology 13 (1), pp. 3-21.
Suvarierol, S. (2007) Beyond the myth of nationality: a study on the networks of European Commission officials, Delft: Eburon.
Versluis, E. (2003). Enforcement matters: enforcement and compliance of European directives in four member states, Delft: Eburon.
32
Walker, D. & Myrick, F. (2006). Grounded theory: an exploration of process and procedure, Qualitative health research 16 (4), pp. 547-559.
Wilson, H.S. & Hutchinson, S.A. (1996). Methodological mistakes in grounded theory, Nursing research 45 (2), pp. 122-124.
33
BACKGROUND OF THE AUTHORS
Lars Tummers (corresponding author) Drs. Lars Tummers, MSc, is studying for a PhD in public administration at the Erasmus University of Rotterdam. His PhD project focuses on the way professionals experience new policies, and uses the innovative concept of 'policy alienation'. Lars also works as a change management consultant at PwC (PricewaterhouseCoopers) Advisory. He has studied both business economics and public administration. [email protected] Address: Erasmus University Rotterdam, Department of Public Administration, Room M7-07, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands
Research Interests Public professionals (New) Public Management Policy implementation Change management Resistance to change Policy alienation Qualitative methodology Quantitative methodology
Niels Karsten Niels Karsten MSc MA, has studied both the philosophy of science and public administration. As a PhD student, he is currently investigating public leadership
34
accountability in consensual democracies at the Tilburg School of Politics and Public Administration (Tilburg University, The Netherlands) [email protected] Address: Tilburg University, TSPPA, Room M3-06, P.O. Box 90153, 5000 LE Tilburg, The Netherlands
Research interests Local government Citizenship Democracy Accountability Qualitative methodology Philosophy of science
ACKNOWLEDGEMENTS The authors would like to thank the two anonymous reviewers plus dr. H.R. Boeije, prof. dr. Mirko Noordegraaf, prof. dr. F.P.J. Wester, and dr. K.J. Mulderij for their constructive comments on earlier versions of this article. The authors are further appreciative of prof. dr. M.A.P. Bovens, dr. S. van Thiel, dr. M.J. van Hulst, and dr. S. Suvarierol for their help in finding the examples of public administration research discussed in this article.
35
i
Naturally, the grounded theory methodological field can overlap with the sector-specific field of
public administration.
ii
For a discussion see: Bryant & Charmaz, 2007; Charmaz, 2006: 165-172; Cutcliffe, 2000: 1480-
1481; Heath, 2006; Heath & Cowley, 2004; Glaser, 1998 and Strauss, 1987.
iii
Please note that we do not need to take sides in this debate. We recognize that using literature
can be, and often is, an essential element of good scientific research. The notion of a blank slate is a misguided and dangerous illusion, even for grounded theorists, and neither Glaser nor Strauss advocate it in the sense of denying prior knowledge or believing in so-called 'pure induction' (see Bulmer, 1979; Charmaz, 1990). Further, we would not claim that the truth lies somewhere between Glaser and Strauss since 'the' truth is not the issue here.
iv
While this division makes the research process appear linear, this in not necessarily the case,
particularly in studies that adopt a GT approach.
v
References inside quotation marks do not appear in the reference list since they are part of the
quote from an original article.
36
doc_616435139.docx
Technology (from Greek τέχνη, techne, "art, skill, cunning of hand"; and -λογία, -logia[1]) is the making, modification, usage, and knowledge of tools, machines, techniques, crafts, systems, and methods of organization, in order to solve a problem, improve a preexisting solution to a problem, achieve a goal, handle an applied input/output relation or perform a specific function.
Research Reports in Reflecting on the Role of Literature in Qualitative Public Administration
ABSTRACT When undertaking qualitative research, public administration scholars must walk a thin line between being theoretically sensitive and imposing preconceived ideas on their work. This article identifies opportunities and pitfalls in using literature in qualitative public administration research. Whereas the opportunities are already well known within the discipline, the pitfalls remain underexposed. We identify potential pitfalls by using insights from the grounded theory approach. To illustrate how opportunities can be optimally exploited, and pitfalls avoided, we provide examples of high quality public administration research. Finally, we derive recommendations for public administration scholars when using literature in their qualitative research. These recommendations can help improve qualitative methods in the public administration discipline.
2
1
PUBLIC ADMINISTRATION RESEARCH AND THE ROLE OF LITERATURE
Textbooks on methods for public administration research contain many arguments in favor of using literature in the research process (e.g. McNabb, 2002:67-68; Bearfield & Eller, 2007:61-63; Justice, 2007:83-84). A literature review of publications that deal with the content of a research field, that is non-methodological literature, such as the key concepts, existing theories, theoretical frameworks, and/or empirical results, serves many purposes. For example, it helps researchers formulate a framework for analyzing their topics and problems (Justice, 2007:83). Further, the literature "provides clarity on a given subject by revealing long standing conflicts and debates, reveals the interdisciplinary nature of research on a subject, and places the work in a historical context" (Bearfield & Eller, 2007:62). Using the literature can have many advantages. However, contrary to what many public administration scholars believe, the overall advantage of using literature in all phases of the qualitative research process is not as self-evident as they suppose because it also comes with potential disadvantages. For instance, an extensive literature review may form a barrier to developing fresh insights, as a researcher may impose "predetermined understanding and existing frameworks on the investigation" (Heath, 2006:519). Further, researchers may select cases that accord with the existing literature and avoid cases or data that may provide contradictory evidence (Heath, 2006:523). If this occurs, then existing theory will have structured the process of data gathering in an undesirable way (Becker, 1993). Hence, questions related to when and how to use literature during the research process are important methodological issues. The issue of when and how to use the literature in qualitative research is relevant to quite a number of social science disciplines. Nevertheless, we argue that it should be
3
of particular interest to public administration scholars for two reasons. Firstly, although there are potential disadvantages in using literature in all qualitative research, it is especially difficult to find arguments against its use in the public administration discipline, or guidelines on how to deal with the dangers of excessive use. Handbooks on methods for public administration rightly indentify the literature review as a crucial step in the design and conduct of research (e.g. McNabb, 2002:393), but they generally fail to recognize the importance of the methodological considerations related to timing and implementation surrounding it. As such, the pitfalls of using literature remain underexposed. Secondly, when and how to use the literature in qualitative research carries particular importance for the discipline of public administration since the quality of qualitative research in this field has been called into question. Brower, Abolafia, and Carr (2000:363), for example, state with regard to public administration as a discipline that "the field's use of qualitative methods reveals substantial weaknesses". They offer four general guidelines for improving qualitative methods in public administration research: "push description to explanation", "improve the quantity of data", "develop theoretical sensitivity", and "develop sensitivity for the backstage" (Brower et al., 2000:387-390). Under the third guideline, they explicitly recommend public administration scholars to "exercise caution to allow the regularities and anomalies in the data to suggest possible theories, rather than force data into theories in ways that obtain a premature closure of meaning" (idem, 389). We elaborate on this recommendation by focusing specifically on the use of the literature and by developing specific lessons for the qualitative public administration research process. The way one deals with the literature plays a crucial and guiding role in the iterative process between theory and data because it determines how existing theories are dealt with in the design and conduct of research. As such, considerations as to when and how to use the literature acquire great importance. A
4
more thorough reflection on the role and use of literature in research in the public administration field may consequently improve the reception and application of public administration research, the legitimacy and usefulness of which according to some remain questionable (e.g. Bogason & Brans, 2008:92). These considerations justify paying significant attention to the role of literature in public administration research. To determine guidelines for how to deal with this important issue, one has to look beyond the public administration discipline. Many of the more critical scholars come from the fields of health and nursing studies. They often use the grounded theory approach, which is very critical of the use of literature (Glaser & Strauss, 1967).i According to Glaser (1998:67), a strong dictum within grounded theory is "do not do a literature study in the area and related areas where the research is to be done". Morse (2002:295), for example, states that the "fear of invalidating one's own work by violating the principles of induction is prevalent in the writing of qualitative researchers. In fact, whether or not to go to the library before beginning fieldwork has become a researcher's dilemma". This somewhat extreme view is often disputed by grounded theorists but, at the very least, many scholars do not take the use of literature for granted (e.g. Becker, 1993; Cutcliffe, 2000; Morse, 2002; Heath & Cowley, 2004). We draw insights from the grounded theory approach as our starting point, and use these to identify the potential pitfalls when using literature, even when using qualitative approaches other than grounded theory. The potential opportunities are derived from both grounded theory and from the public administration literature. We explicitly choose to speak in terms of opportunities and pitfalls, rather than, for example, in terms of advantages and disadvantages. This is because we believe that a good researcher is able to minimize the disadvantages of a particular method and to fully exploit its advantages. From this overview of pitfalls and opportunities, we derive
5
recommendations for public administration scholars when using literature in qualitative research. Our focal question is:
What are the opportunities and pitfalls when using literature, and what recommendations for public administration scholars undertaking qualitative research can be derived from these?
The outline of this article is as follows. First, we introduce grounded theory and consider its perspective on the use of literature. Second, we identify a number of opportunities and pitfalls. We illustrate these by providing concrete examples taken from public administration research that deal with the literature effectively. Third, we formulate recommendations for public administration scholars to follow when using literature in qualitative research. 2 THE GROUNDED THEORY APPROACH
In this section, we outline the general characteristics of 'the' grounded theory approach. A difficulty is that, since the methodology was first developed in the 1960s, two very different approaches have evolved, namely those of Glaser and of Strauss. The methodological differences between these two approaches are substantial. Some academic scholars even argue that two fundamentally different research strategies have evolved (Glaser, 1992:2; Wilson & Hutchinson, 1996; Cutcliffe, 2000:1481 ff.). However, we do not share this view. What has evolved, we would argue, are two methodological approaches that belong to the same research strategy. The two approaches share essential defining characteristics that make them grounded theory approaches (see also Benoliel, 1996; Melia, 1996; Cutcliffe, 2000; Walker & Myrick; 2006:549-550; Bryant &
6
Charmaz, 2007). Before going into the detail of the grounded theory approach, we provide the background to its development and its relationship with other approaches.
2.1
Background of the grounded theory approach
The grounded theory approach was developed in the 1960s by Barney Glaser and Anselm Strauss (Glaser & Strauss, 1967; see also Boychuk Duchscher & Morgan, 2004:606). In those years, qualitative sociological research was losing ground to more quantitative methods. This shift was closely related to the increasing dominance of the positivistic paradigm. Charmaz (2006:4) notes that "only narrowly scientific - that is, quantitative - ways of knowing held validity for positivists; they rejected other possible ways of knowing, such as through interpreting meanings or intuitive realizations". In their book The discovery of grounded theory, Glaser and Strauss (1967) questioned those epistemological assumptions. They made an innovative statement, contesting those notions and at the same time offering systematic qualitative research strategies (Charmaz, 2006). In essence, they argued that research flowing from the positivistic stance had led to the development of abstract theories that no longer did justice to empirical realities. As an alternative, they proposed a method for systematic qualitative analysis that would generate theories that were 'grounded' in empirical realities. The impact of their book on qualitative research has been substantial. It was firstly influential in nursing and health studies, most prominently those undertaken at the University of California. Over the years, scholars in many other fields, such as public management and the sociology of professions, have embraced this research approach (Charmaz, 2006). For instance, today, professionals involved in student affairs use
7
grounded theory to increase educators' understanding of the complex interactions between students and college environments (Bowen, 2008). Examining its relationship with other research methods, grounded theory can be positioned among other inductive, qualitative approaches (Strauss & Corbin, 1990). In inductive analyses in general, patterns, themes, and categories of analysis come from the data rather than from predetermined theory. Theory is derived from the data, rather than being imposed on the data prior to data collection or analysis (Patton, 1980). That is, a grounded theory is derived inductively, through the systematic collection and analysis of data using a continuous comparative approach. Related approaches - which do differ from grounded theory in a number of aspects - include phenomenology and ethnography. At the other end of the spectrum, there are distinctly different approaches such as deductive, qualitative or quantitative, analyses. Deductive approaches start from a theory and subsequently test this theory using empirical analyses. Grounded theory explicitly runs counter to these more deductive approaches, especially to those that use quantitative methods.
2.2
Characteristics of the grounded theory approach
Having provided a short background to grounded theory, we will now describe its main characteristics. The first important characteristic of the grounded theory approach has already been mentioned: grounded theory is inductive (Eaves, 2001:655; Heath & Cowley, 2004:142). The second key characteristic is that the grounded theory approach has a specific way of developing theories, namely through "purposeful systematic generation from the data of social research" (Glaser & Strauss, 1967:28). Theories are constructed from the data themselves (Eaves, 2001:655-656). The supposed advantage
8
of this is that the theories that are being developed, almost by definition, are grounded in the data and therefore do justice to the social reality. The third general characteristic of grounded theory is that it aims to develop explanatory theories which, as such, are very different from descriptive or purely conceptual theories. Alongside these three general characteristics, two shared grounded theory characteristics can be identified that relate to the methodology. The first methodological characteristic shared by both grounded theory approaches is that of constant comparison. During a grounded theory study, data are continually compared with: 1) other primary data, 2) evolving original data, 3) evolving concepts, and 4) evolving theories (see Boychuk Duchscher & Morgan, 2004:607; see also Glaser & Strauss, 1967). In making these comparisons, grounded concepts and the relationships between them become visible. The second methodological characteristic shared by both grounded theory approaches is the process of 'open coding' of data. By assigning codes that have not been fixed in advance to data during the analysis, and by a continuous comparison between codes, grounded theorists develop concepts that form the basis of evolving theories. As already noted, two distinct grounded theory approaches have evolved since the 1990s. The differences between the Glaserian and the Straussian approaches spring from an intense debate between the two on what grounded theory should look like. In 1990, Strauss, in cooperation with Juliet Corbin, published Basics of Qualitative Research. This book is at the heart of the debate between Glaser and Strauss (Eaves, 2001:656). Strauss introduced a new coding method: "a coding paradigm involving conditions, context action/interactional strategies and consequences" (Strauss & Corbin, 1990:67). Even though both approaches adhere to the principle of 'open coding', the differences between the respective coding methods of Glaser and of Strauss are substantial (see Strauss & Corbin, 1990; Kendall, 1999; Eaves, 2001; Boychuk
9
Duchscher & Morgan, 2004; Glaser & Holton, 2004; Heath & Cowley, 2004; Walker &
Myrick, 2006). Rather than letting theory emerge from the data, Strauss 'forces' the data and the theory into a methodological framework which, it is argued by critics of this approach, violates the principle of induction (see also Boychik Duchscher & Morgan, 2004). According to Glaser, this has led Strauss away from the basics of grounded theory and toward a method that can no longer be considered as grounded theory (Glaser, 1992:2). 2.3 The use of existing literature in grounded theory
One aspect of the debate between Glaser and Strauss concerns the question as to when and how literature should be used in the research process. In this article, this is an important topic as it is closely linked to our aim, which is to see what lessons can be learnt for public administration research from the debate among grounded theorists on the use of literature in qualitative research. First and foremost, it is well worth noting that the role of literature is relatively minor in both Glaser's and Strauss' approaches (Baker et al., 2002). Essentially, the ideas underpinning grounded theory are very critical of its use, and both approaches remain loyal to this sentiment. Since grounded theory is designed to inductively derive theories from the data, existing literature should be viewed with suspicion at the very least. Grounded theorists generally fear that they will violate the principle of induction as a consequence of using literature either too early or in the wrong way (e.g. Glaser, 1998:68-69). This is not an imagined fear; there is a real danger that a researcher will violate one of grounded theory's main principles through using literature: theories should be derived from the data, and from nothing else. However, most grounded theorists do recognize the usefulness of literature in at least some phases of a research project, and
10
do try to incorporate its use in their methods, albeit in very different ways (Heath & Cowley, 2004:143). In essence, Strauss 'allows' the existing literature to function as a framework for analysis early in the research process; whereas Glaser only uses the literature later, and then only in the role of providing comparative data. Here, for two reasons, we will not discuss in detail the differences in use of the literatureii, but instead focus on the arguments that are being advanced in the debate. Our first reason is that the place and function of a literature review in Glaser's and Strauss' approaches are intensely debatediii and not unambiguous (Charmaz, 2006:165). Second, we are not so much interested in the methodological aspects of grounded theory, but more in the arguments that can play a role in the ongoing debate about the use of literature in grounded theory. Clearly, this 'methodological battle' is far from over. Nevertheless, the methodological literature on grounded theory provides a valuable collection of arguments for and against the use of literature in different phases of grounded theory studies. Our aim is to see what lessons can be learnt from this debate that could be applied to other qualitative research strategies that are used in the field of public administration. We believe that the arguments advanced could function as a valuable mirror for public administration scholars. Both Glaser and Strauss, and their respective followers, provide insights that, while they primarily apply to grounded theory methodology, can be transposed in such a way that they shed light on the opportunities and pitfalls of using literature in public administration research in general. Using Glaser's and Strauss' arguments as a guide, this is what makes grounded theory such a useful perspective in more generally reflecting on the role of literature in qualitative research. What is innovative about our approach is that it explicitly combines both grounded theory approaches, whereas previous reflections on the role of literature have, either explicitly or implicitly, tended to take either a Glaserian or a Straussian perspective and, furthermore, were usually limited to a grounded theory perspective. We believe that the
11
two perspectives can be combined, and that there are also lessons to be drawn that are relevant for public administration scholars. 3 OPPORTUNITIES AND PITFALLS IN USING LITERATURE DURING EACH SUCCESSIVE RESEARCH PHASE
In this section, we begin to identify the opportunities and pitfalls in using literature. This discussion is not so much focused on the question as to whether researchers should read and review the literature: without doubt, researchers have to relate their work to the existing literature in their field. The opportunities and pitfalls are more concerned with when to use literature, and if and how to use it in analysis. Glaser and Strauss both argued for delaying, rather than abandoning, the literature review (Charmaz, 1990:1163). Grounded theory can warn us of the opportunities and pitfalls in reviewing literature early in the research project, i.e. before the initial analysis has been completed (see also Glaser, 1998:67-80). Given that the opportunities and pitfalls are (partly) related to when one uses the literature, we have categorized them according to the research phase in which they are most likely to occur. We distinguish three main phases, which we will discuss separately below: research design (3.2), data collection (3.3), and data analysis (3.4).iv 3.1 Selecting illustrative public administration examples
The opportunities and pitfalls we have identified can remain quite abstract. Throughout this section we therefore discuss research drawn from the public administration discipline that effectively used - or deliberately did not use - a literature review in the various phases of the research process. For each opportunity, and for each pitfall, we give an example of good quality public administration research and we briefly explain
12
how the researcher minimized the effects of the applicable pitfall, or exploited the opportunity. Note that we do not consider these examples to be best practices. Rather, the selection includes those studies which we believe to be the most instructive for the reason that the authors provide an elaborate account of the considerations behind the way in which they used literature. We hope doctoral students and established scholars find inspiration in these examples, which could function as illustrations of how to handle literature reviews in qualitative research. We used four methods to gather examples. First, we searched scientific databases. As different databases scan different sets of journals, we used a number of databases: ABI/Inform Global, JSTOR, Sage Journals online, and ISI Web of Knowledge. Topics that were used in searching these databases included [public administration], [political science], [policy studies], [qualitative], [inductive], [grounded theory], [deductive], [literature] and [literature review], as well as a number of synonyms. Second, we analyzed the abstracts of articles published between 2005 and June 2009 in five leading public administration journals (Administration & Society, Journal of Public Administration Research & Theory, Public Administration, Public Administration Review, and Public Management Review). Third, we used the descendancy approach (Bowen & Bowen, 1999:74), which amounts to locating all studies cited in key publications. We used Brower et al. (2000) as our key publication. Fourth, we asked fifteen Dutch public administration scholars, five of whom can be considered to be methodological experts, to provide good quality research examples for each of the identified opportunities and pitfalls. Four of them provided suitable examples. We used four main selection criteria. First, we only selected research from the field of public administration. Second, we only chose studies that took a qualitative approach. Third, only publications in English were considered. Fourth, the research
13
report had to pay explicit attention to considerations regarding the use of literature. The selections made of this basis include those studies we believe to be the most instructive. Note that not all authors were claiming to be using grounded theory, with the notable exceptions of Agranoff (2007), DeHart-Davis (2008), and De Kort and Van der Pijl (2009). As such, we are also drawing on other research strategies, since these can also provide valuable insights into the use of literature. 3.2 Phase 1 - Research design
In the initial research design phase, the researcher formulates a research question, selects a methodological approach, and then operationalizes the relevant variables (McNabb, 2002). In this phase, we have identified three opportunities and one pitfall related to literature use. A first major, and well known, opportunity is that reading literature helps researchers identify 'knowledge gaps' (Hutchinson, 1993). Subsequently, it becomes easier to connect a research project to other research, to show the theoretical relevance of a study, and to integrate the evolved theory with other theories. The added value of a study, that is its contribution to the body of knowledge, hence increases. Grasping this opportunity can be illustrated using Ackroyd, Kirkpatrick, and Walker's study of public management reform in three services in the UK: healthcare, housing, and social services (2007). In their article, they note that "In the literature there are now many overviews of this process [of public management reform], but few systematically compare the effects of restructuring. With some exceptions (see Boyne et al., 2003)v accounts tend either to analyze public services as separate cases or emphasize general trends in them all. Hence, what is often lacking is a 'comparative analysis across different domains' (Ferlie et al., 2003:S3)". Here, Ackroyd and his coauthors use the existing literature to identify a knowledge gap, arguing that comparative
14
research in their field is lacking. Their study then starts to compare public management reform in three UK services. Through this, they add theoretical relevance to their study. If they had not used existing literature, it would have been more difficult for them to connect their project to other research, and it could be the case that they would have researched something which was already well-known within the discipline. Secondly - and partly related to the first opportunity - the use of literature can help to explain to the reader why a particular question is important. It helps in answering the 'so what' question. Although a question may sound interesting, or provoke curiosity, to properly establish its importance, research must be placed in context. A wellconstructed literature review can provide a panoramic view of the research question and help the reader understand why a particular question matters (Bearfield & Eller, 2007:62). This can be illustrated using the scholarly work of Esther Versluis Enforcement Matters: Enforcement and compliance of European Directives in four Member States (2003). This is a comparative analysis, involving four member states of the European Union, of the transposition and enforcement of, and compliance with, two directives regulating dangerous chemical substances. With reference to existing literature, she shows that a comparative study of compliance and enforcement is important. She for example states that "Strict regulation can lose its meaning when enforcement agencies keep an eye closed when monitoring companies or when they make no use of sanctions. Obedience differs between countries - 'In some countries, rules are there to be followed, in others it seems if they merely exist to be violated.' (Van Waarden, 1998:2) - and insufficient monitoring of the enforcement process by the Commission probably leads to diverging enforcement practices within the Member States." (Versluis, 2003:10). She then continues to formulate her research questions on this topic of possible diversity in compliance and enforcement. Hence, she is placing the research into context and
15
highlighting why the particular research questions are important. If she had not referred to existing literature to highlight why a research question was relevant, it would have been far more difficult to answer the 'so-what' question: why does this research matter, and how does it add to existing knowledge? Hence, the way she used existing literature provided legitimacy for conducting large-scale research on comparing compliance and enforcement in a number of EU member states. Finally, using the existing literature can help focus a research project (McNabb, 2002:67). Literature can significantly narrow the focus of a research project by providing tentative questions or enhancing conceptual clarity. Cutcliffe, (2000:1480), using his own doctoral study on hope as an example, notes that "Many proposed research questions require conceptual clarity. [A literature] review may help provide a sense of the key elements of hope that are implicit in the literature, it may help provide some conceptual clarity of the nature of hope and the nature and practice of bereavement counseling and this examination of the relevant literature would help the researcher to differentiate hope from similar and related concepts." In their recent JPART article, Foldy and Buckley (2010) examine team learning among street-level bureaucrats. State-of-the-art literature from the fields of social psychology, organizational behavior, and public administration was arguing in favor of focusing on the group level, rather than for instance on individual characteristics, if one is to effectively examine team learning. Therefore, the authors of this article decided to examine teams, documenting whether they discarded old routines and learned new ones. Also in other phases of their research, Foldy and Buckley often used literature to narrow their focus. They, for example, constructed hypotheses based on scholarly work: "Drawing on previous research on both team effectiveness and learning, we developed hypotheses on the team characteristics that could have an impact on team learning." (ibid:25). It is worthwhile considering what could have happened if they had not used the
16
existing literature to provide focus. If they had not focused on the group level, this could have resulted in a less clear-cut article, maybe one looking at individual characteristics, organization characteristics and team characteristics. This would have provided few insightful additions to the scholarly literature and for practitioners. However, using literature in this phase of research also has a substantial pitfall. According to Becker (1993) and Heath and Cowley (2004), a researcher runs the risk of overlooking social phenomena, or relevant aspects, by focusing exclusively on those issues that seem relevant according to the literature. Referring to the work of Glaser (1998:67-73), Heath and Cowley (2004:143) note that researchers should be aware of "near misses in discovering new theory, [which] is a process whereby as the theory begins to emerge, literature of close relevance is recognized or read and its powerful impact bends the emerging theory from its true path." This can also be illustrated using an example from the public administration discipline. According to numerous EU studies, nationality forms a basis for networks in the European Commission. However, authors do not question this relationship, and therefore do not systematically assess this assumption. Suvarierol's book Beyond the myth of nationality (2007) goes further and challenges the current EU literature by systematically exploring the role of nationality and networks in the European Commission. In one important section, entitled 'Debunking the myth of nationality', Suvarierol concludes that "The results can be summarized as follows. Even though many officials recognize the mechanisms that underlie homophily resulting from cultural similarities, the networks of officials are not culturally homophilic." (ibid:104). Essentially, she is stating that the networks of these officials are not as nationally or culturally oriented as is often assumed. In her book, Suvarierol focuses on other social phenomena which are generally not considered relevant in the literature, thereby avoiding the pit that others had been steered into. If, as other scholars, she had
17
accepted this nationality myth at face value, she would most likely have missed important ways in which EU officials work in Brussels. She would have simply assumed that national networks play a major role, and maybe would have interpreted her data using this assumption. Consequently, examining this often accepted phenomenon enabled Suvarierol to debunk some common myths about the way officials work in the European Union. 3.3 Phase 2 - Data collection
In the data collection phase, data are gathered in order to satisfy the research objectives and answer the research question (McNabb, 2002:56). An opportunity in using literature in the data collection phase is that it can provide an additional data source, alongside, for example, interviews and observations. This corresponds with Glaser's notion that "all is data" (Glaser, 2001:145). In this way, the researcher can gain a greater understanding of the relevant phenomena. For the purposes of this article, it is unfortunate that the number of public administration studies that include existing literature as another data source is low. One of the few exceptions from SSCI ranked journals that we found is a 2009 article by De Korte and Van der Pijl. The authors included existing articles in both the data collection and the data analysis phases of their study on organizational change within the setting of a governmental bureaucracy. Regrettably, the role of these articles is not specified in great detail, which makes it difficult to assess the effect of including existing literature as another data source in the study. Our interpretation is that the additional 'data' enriched their analysis by providing codes that could be compared with other codes that emerged from other data. On the other hand, using literature in this phase has a major pitfall. Researchers may go on to select cases (or data within cases) for their research that accord with the
18
existing literature and avoid cases or data that may provide contradictory evidence (Heath, 2006:523). If this occurs, existing theory will have structured the process of data gathering in an undesirable way (Becker, 1993). The end result of such a data collection process may be a biased picture - one that does not correspond with reality. In his study of so-called 'public management networks', Agranoff (2007) used a grounded theory approach because he "had to dig deeper into the data, looking for a form of meaning that might lie outside the previous literature" (ibid:36). In his view, he needed to go "beyond the trendy assertions" (ibid:35) of existing, well-established network theories that held that networks functioned as decision-makers. Agranoff argued that although existing theories could function as value frameworks for analysis, they lacked a perspective on those networks that do not make decisions. In an alternative approach, Agranoff set out to uncover network management as practiced by its practitioners. If, in the data collection phase of the research, he had only concentrated on data that were relevant according to the existing literature he would most likely have missed important roles of public management networks other than decision-making, because the existing literature provided little room for these. Subsequently, grounded theory enabled Agranoff to generate the contextualized knowledge that existing network theories were lacking in his view (see also Bogason & Brans 2008:92-93). 3.4 Phase 3 - Data analysis
Once the data are available, the researcher has to order the data and determine their meaning. This interpretation is necessary because it enables the findings to be related to the original research question and research objectives (McNabb, 2002:57-58). In this data analysis phase, one opportunity gained from using literature is that researchers can enhance their insights into the object under study (Cutcliffe, 2000:14801481; Strauss & Corbin, 1990). For example, an insight may be enhanced through the
19
literature increasing a researcher's sensitivity to latent relationships within the data: that is, it strengthens theoretical sensitivity (Strauss, 1987; Heath & Cowley, 2004:143; see also Brower et al., 2000:389). In this way, prior comprehension plays a positive role in analyzing the new data (Heath, 2006). This helps the researcher to achieve the main goal of qualitative research: "Verstehen" (understanding). In 1971, Allison published his famous analysis of the Cuban missile crisis. Instead of only using rational expectancy theories, Allison used two additional models to explain how the US and the Soviet Union made decisions in this Cold War event. The organizational behavior model and the government politics model enabled him to enhance his insights, because each model focused on particular relationships and concepts that were more-or-less absent in the other models. The theoretical models that Allison derived from the existing literature functioned as supplementary frames of reference (Allison & Zelikow, 1999:5), providing him with additional interpretations of decision-making processes. This approach provided a much more all-encompassing view on how decisions were made. This research strategy enabled Allison to draw on existing literature in order to understand what had happened. If he had not used the results of his literature review in the analysis, he would not have benefited to the same extent from such well established framework models, including their core theoretical concepts that increased his sensitivity to latent relationships within the data. Conversely, using literature in this phase also has a pitfall. It can form a 'barrier of insight' (Heath, 2006:520; Glaser, 1978), with the result that a researcher interprets data through a framework provided by the literature. Consequently, a researcher may impose "predetermined understanding and existing frameworks on the investigation" (Heath, 2006:519, see also Glaser, 1978; Cutcliffe, 2000:1481; Stern, 1994). In other words, researchers import preconceived ideas - such as existing theoretical structures and impose them on their own work (Charmaz, 2006:165; Charmaz, 1990:1163; cf.
20
Glaser, 1998:67 ff.). This can suppress fresh insights that might otherwise have emerged from the data. DeHart-Davis (2008:363-364), for example, feared that the extensive body of literature on the nature and consequences of ineffective rules, or 'red tape', would form a barrier of insight in her study on effective organizational rules, dubbed 'green tape'. The red tape literature, in her view, had a strong presumption that rules inhibit creativity and flexibility. Further, it is said to suffer from a disproportionate focus on middle-level and senior managers. A study of 'green tape', DeHart-Davis argues, consequently runs the risk of anchoring respondents' perceptions of rule quality in pre-existing theory. To avoid this danger, she avoids using existing theories as a framework for her analysis, and instead identifies patterns in the data through a grounded theory approach. Had she used the existing literature as a starting point for analysis of her interview data, it would have been difficult to detect the attributes of green tape because the literature provides little room for such rules. It was for that reason that DeHart-Davis employed a grounded theory research strategy which allowed her to indentify and categorize the attributes of green tape without imposing preconceived ideas of what are 'good' or 'bad' rules on her work. For another good example of avoiding barriers of insight, see Durant's (2007) study on large-scale organizational change in public organizations. He conducted extensive, inductive archival research in an "understudied arena of public policy" (idem, 410) and this strategy enabled him to avoid imposing preconceived ideas on his work. 3.5 Summarizing opportunities and pitfalls
Based on the discussion above, we have constructed Table 1 which summarizes the opportunities and pitfalls in each research phase, and notes the related examples drawn from public administration literature discussed above.
21
Opportunities/pitfalls in each research phase
Public administration literature which used the opportunity optimally or effectively to avoid the pitfall
1. Research Design a. Opportunity: identifying knowledge gaps Ackroyd, Kirkpatrick, & Walker (2007) b. Opportunity: explaining why a particular research question is relevant ('answering the 'so what' question) c. Opportunity: focusing the research project Foldy & Buckley (2010) Suvarierol (2007) Versluis (2003)
d. Pitfall: Overlooking social phenomena
2. Data collection a. Opportunity: using literature as an additional data source b. Pitfall: selecting only data that are relevant according to the literature Agranoff (2007) De Korte & Van der Pijl (2009)
3. Data analysis a. Opportunity: enhancing insight by increasing sensitivity to latent relationships b. Pitfall: forming a barrier to fresh insights DeHart-Davis (2008), Durant (2007) Allison (1971)
22
4
RECOMMENDATIONS FOR PUBLIC ADMINISTRATION RESEARCH
We have detailed a number of opportunities and pitfalls related to using literature in new research. These have been established by drawing on literature on both grounded theory and public administration research. Further, we have provided examples from existing public administration research that demonstrated each of the opportunities and pitfalls. The main point being emphasized is that a benefit is not inevitable from using literature: there are both opportunities and pitfalls connected to using it. On this basis, we argue that public administration scholars should take a more critical stance toward the use of literature reviews in the various phases of the qualitative research process, and that grounded theory can provide inspiration in this direction. This conclusion has a number of consequences. First, we distinguish two general recommendations and we follow these by more-specific recommendations for the various stages of the research process.
General recommendations: The question as to whether to use literature or not deserves consideration in each phase of the qualitative research process. The use of literature in all research phases is not by definition a defining characteristic of sound research.
Recommendations for research design: In the research design phase, public administration scholars should indicate and justify where literature will be used in each of the remaining research phases, in
23
what role, and to what extent. Ideally, this should primarily depend on the research questions and the research goals. A thorough research design should further indicate how the potential pitfalls of using literature will be avoided and how the opportunities will be utilized to their full potential.
Recommendations for data collection and analysis: During data collection and analysis, the role of literature and its influence on the development of the research, and its results, should be continuously reviewed.
Recommendations for research reporting: Research reports should be explicit as to the phases in which literature has been used, and to what extent and how the opportunities and pitfalls have been addressed, and what limitations follow from this. Academic publications should similarly be explicit about how literature has been used during the research process, as well as about the limitations this imposes on the results.
To sum up, we have argued that explicitly questioning the role of literature is an essential part of the qualitative research process. If public administration scholars were to do this systematically, we believe that it would substantially enhance the quality of this research field. In our view, this applies to the diverse set of qualitative research approaches adopted in the public administration field and, as such, moves beyond grounded theory. It could strengthen the theory building capacity of the various research approaches, such as interpretivism, empiricism, rationalism, and postpositivism, used in the public
24
administration discipline as discussed by Riccucci (2010:65-96). She argues that all these forms of research could potentially contribute to theory building in public administration. We believe that these various research approaches can learn - to a greater or lesser extent - from recommendations derived in this article. This is because we do not generate specific do's and don'ts (which could be incompatible with some approaches) but, instead, encourage a more critical stance and awareness of the potential pitfalls in using literature in general. As such, we promote a continuous review of its use throughout the various research stages, and remaining aware and explicit about the choices that follow from that review. We end this article with a final, more practical, concern. One might worry that there may well be obstacles to pursuing these recommendations in terms of journal articles and research funding because, throughout the public administration discipline, literature reviews are usually regarded as an essential part of all research projects (see also Fink, 2005:6). This is an honest observation and, more generally, Schreiber (2001:57) argues that:
"The current expectations of academic research and funding agencies suggest that plunging into field research without delving into the relevant literature would be folly. [?] The researcher who is seeking funding must demonstrate an understanding of the 'state of the science' regarding the phenomenon of study in order for agency evaluators to be confident that their money would be wellspent."
These practical concerns can still be aligned with our recommendations as our message is not that literature reviews should not to be included in research designs or research reports. It is rather that the researcher should include a thorough explanation and
25
justification for the use of literature in the various phases of the research project. This should improve rather than diminish the quality of public administration research since it provides an underpinning to what should be a critical methodological consideration. Nevertheless, what will be required is a willingness by reviewers to fund and publish research that delays elements of the literature review until an appropriate research phase, recognizing that this may not necessarily be the design phase, provided that the researcher provides a sound methodological justification for the timing. This fits with a trend in which a growing number of public administration scholars employ interactive research strategies. According to Bogason and Brans (2008:92-93), this will help overcome the alleged cool reception and limited application of public administration theory in practice. Such developments should make funding, as well as publication, of public administration research easier.
26
REFERENCES
Ackroyd, S., Kirkpatrick, I. & Walker, R. M. (2007). Public management reform in the UK and its consequences for professional organization: A comparative analysis. Public administration, 85 (1), pp. 9-26.
Agranoff, R. (2007). Managing within networks: adding value to public organizations, Washington: Georgetown University Press.
Allison, G. & Zelikow, P. (1999). Essence of decision: explaining the Cuban missile crisis, Nw York: Longman
Baker, C., Wuest, J. & Stern, P.N. (2002). Method slurring: the grounded theory phenomenology example, Journal of advanced nursing, 17 (11), pp. 1355-1360.
Bearfield, D.A. & Eller, W.S. (2007). Writing a literature review: the art of scientific literature, in: Yang, K. & Miller, G.J. (eds.). Handbook of research methods for public administration, New York: CRC Press, Taylor & Francis, pp. 61-72.
Becker, P.H. (1993). Common pitfalls in published grounded theory research, Qualitative health research 3 (2), pp. 254-260.
Benoliel, J.Q. (1996). Grounded theory and nursing knowledge, Qualitative health research 6 (3), pp. 406-428.
27
Bogason, P. & Brans, M. (2008). Making public administration teaching and theory relevant, European Political Science, 7, p. 84-97
Boychuk Duchscher, J.E. & Morgan, D. (2004). Grounded theory: reflections on the emergence vs. forcing debate, Journal of advanced nursing 48 (6), pp. 605-612.
Bowen, G.A. (2008). Grounded theory and sensitizing concepts. International Journal of Qualitative Methods, 5(3), pp. 12-23.
Bowen, W.M. & Bowen, C.C. (1999). Typologies, indexing, content analysis, metaanalysis and scaling as measurement techniques, in: G. Miller & M. Whicker (eds.) Handbook of quantitative methods in public administration, New York, NY: Marcel Dekker, pp. 51-86.
Brower, R.S. Abolafia, M.Y. & Carr, J.B. (2000). On improving qualitative methods in public administration research, Administration & Society 32 (4), pp. 363-397.
Bryant, A. & Charmaz, K. (eds.) (2007). The SAGE handbook of grounded theory, London: SAGE.
Bulmer, M. (1979). Concepts in the analysis of qualitative data, Sociological Review 27 (4), pp. 651-677.
Charmaz, K. (1990). 'Discovering' chronic illness: using grounded theory, Social Science and Medicine 30 (11), pp. 1161-1172.
28
Charmaz, K. (2006). Constructing grounded theory: a practical guide through qualitative analysis, London: SAGE.
Cutcliffe, J.R. (2000). Methodological issues in grounded theory, Journal of advanced nursing, 31 (6), pp. 1476-1484.
DeHart-Davis, L. (2009). Green tape: a theory of effective organizational rules, Journal of Public Administration Research and Theory, 19 (2), pp. 361-384
Durant, R.F. (2007). Toxic politics, organizational change, and the 'greening' of the U.S. military: toward a polity-centered perspective, Administration & Society 39 (3), pp. 409446.
Eaves, Y.D. (2001). A synthesis technique for grounded theory data analysis, Journal of advanced nursing 35 (5), pp. 654-663.
Fink, A. (2005). Conducting research literature reviews: from the internet to paper, Thousand Oaks: SAGE
Foldy, E.G. & Buckley, T.R. (2010). Re-creating street-level practice: the role of routines, work groups, and team learning, Journal of Public Administration Research and Theory, 20 (1), pp. 23-52.
Glaser, B.G. (1978). Theoretical sensitivity: advances in the methodology of grounded theory, Mill Valley: Sociology Press.
29
Glaser, B.G. (1992). Emerging vs. forcing: basics of grounded theory analysis. Mill Valley: Sociology Press.
Glaser, B.G. (1998). Doing grounded theory: issues and discussions, Mill Valley: Sociology Press.
Glaser, B.G. (2001). The grounded theory perspective: conceptualization contrasted with description, Mill Valley: Sociology Press.
Glaser, B. G. & Holton, J. (2004). Remodeling grounded theory, Forum: Qualitative social research 5 (2), article 4, pp. 1-17.
Glaser, B.G. & Strauss, A.L. (1967). Discovery of grounded theory: strategies for qualitative research, Chicago: Aldine.
Heath, H. & Cowley, S. (2004). Developing a grounded theory approach: a comparison of Glaser and Strauss, International journal of nursing studies 41 (2), pp. 141-150.
Heath, H. (2006). Exploring the influences and use of the literature during a grounded theory study, Journal of research in nursing 11 (6), pp. 519-528.
Hutchinson, S.A. (1993). Grounded theory: the method, in: Munhall, P.L. & Boyd, C.O. (eds.). Nursing research: a qualitative perspective, New York: National League for Nursing, pp. 180-212.
30
Justice, J.B. (2007). Purpose and significance of research design, in: Yang, K. & Miller, G.J. (eds.). Handbook of research methods for public administration, New York: CRC Press, Taylor & Francis, pp. 75-92
Kendall, J. (1999). Axial coding and grounded theory, Western journal of nursing research 21 (6), pp. 743-757.
Korte, R. de, & Pijl, G. van der (2009). Characteristics of a central change programme within a governmental bureaucracy: a grounded theory study, Journal of Management & Governance 13 (1-2), pp. 5-40
Melia, K.M. (1996). Rediscovering Glaser, Qualitative Health Research 6 (3), pp. 368378.
McNabb, D.E. (2002). Research methods in public administration and nonprofit management: quantitative and qualitative approaches, Armonk: ME Sharpe.
Morse, J.M. (2002). Theory innocent or theory smart?, Qualitative health research 12 (3), pp. 295-296.
Morse, J.M. & Field, P.A. (1995). Qualitative research methods for health professionals, California: SAGE.
Patton, M.Q. (1980). Qualitative evaluation methods, Beverly Hills: SAGE.
31
Riccucci, N.M. (2010). Public administration: traditions of inquiry and philosophies of knowledge, Georgetown: Georgetown University Press.
Schreiber, R.S. (2001). The "how-to" of grounded theory: avoiding the pitfalls, in: Schreiber, R.S. & Stern, P.N. (eds.), Using grounded theory in nursing. New York: Springer.
Stern, P. N. (1994). Eroding grounded theory, in: J.M. Morse (ed.) Critical issues in qualitative research methods, Thousand Oaks: SAGE, pp. 212-223.
Strauss, A.L. (1987). Qualitative analysis for social scientists, New York: Cambridge University Press.
Strauss, A.L. & Corbin, J.M. (1998). Basics of qualitative research: techniques and procedures for developing grounded theory, London: SAGE.
Strauss, A.L. & Corbin, J.M. (1990). Grounded theory research: procedures, canons and evaluative criteria, Qualitative sociology 13 (1), pp. 3-21.
Suvarierol, S. (2007) Beyond the myth of nationality: a study on the networks of European Commission officials, Delft: Eburon.
Versluis, E. (2003). Enforcement matters: enforcement and compliance of European directives in four member states, Delft: Eburon.
32
Walker, D. & Myrick, F. (2006). Grounded theory: an exploration of process and procedure, Qualitative health research 16 (4), pp. 547-559.
Wilson, H.S. & Hutchinson, S.A. (1996). Methodological mistakes in grounded theory, Nursing research 45 (2), pp. 122-124.
33
BACKGROUND OF THE AUTHORS
Lars Tummers (corresponding author) Drs. Lars Tummers, MSc, is studying for a PhD in public administration at the Erasmus University of Rotterdam. His PhD project focuses on the way professionals experience new policies, and uses the innovative concept of 'policy alienation'. Lars also works as a change management consultant at PwC (PricewaterhouseCoopers) Advisory. He has studied both business economics and public administration. [email protected] Address: Erasmus University Rotterdam, Department of Public Administration, Room M7-07, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands
Research Interests Public professionals (New) Public Management Policy implementation Change management Resistance to change Policy alienation Qualitative methodology Quantitative methodology
Niels Karsten Niels Karsten MSc MA, has studied both the philosophy of science and public administration. As a PhD student, he is currently investigating public leadership
34
accountability in consensual democracies at the Tilburg School of Politics and Public Administration (Tilburg University, The Netherlands) [email protected] Address: Tilburg University, TSPPA, Room M3-06, P.O. Box 90153, 5000 LE Tilburg, The Netherlands
Research interests Local government Citizenship Democracy Accountability Qualitative methodology Philosophy of science
ACKNOWLEDGEMENTS The authors would like to thank the two anonymous reviewers plus dr. H.R. Boeije, prof. dr. Mirko Noordegraaf, prof. dr. F.P.J. Wester, and dr. K.J. Mulderij for their constructive comments on earlier versions of this article. The authors are further appreciative of prof. dr. M.A.P. Bovens, dr. S. van Thiel, dr. M.J. van Hulst, and dr. S. Suvarierol for their help in finding the examples of public administration research discussed in this article.
35
i
Naturally, the grounded theory methodological field can overlap with the sector-specific field of
public administration.
ii
For a discussion see: Bryant & Charmaz, 2007; Charmaz, 2006: 165-172; Cutcliffe, 2000: 1480-
1481; Heath, 2006; Heath & Cowley, 2004; Glaser, 1998 and Strauss, 1987.
iii
Please note that we do not need to take sides in this debate. We recognize that using literature
can be, and often is, an essential element of good scientific research. The notion of a blank slate is a misguided and dangerous illusion, even for grounded theorists, and neither Glaser nor Strauss advocate it in the sense of denying prior knowledge or believing in so-called 'pure induction' (see Bulmer, 1979; Charmaz, 1990). Further, we would not claim that the truth lies somewhere between Glaser and Strauss since 'the' truth is not the issue here.
iv
While this division makes the research process appear linear, this in not necessarily the case,
particularly in studies that adopt a GT approach.
v
References inside quotation marks do not appear in the reference list since they are part of the
quote from an original article.
36
doc_616435139.docx