Qualitative research & academy of management journalDocument Transcript
Academy of Management Journal2004, Vol. 47, No. 4, 454–462. FROM THE EDITORS Qualitative Research and the Academy of Management Journal Editor’s note: For this issue’s “From the Editors,” agement Journal. Qualitative research is importantI invited Robert Gephart of the University of Al- to AMJ. Qualitative research is actively sought andberta to reflect on his observations as a long-serv- supported by the Journal, its editors, and its edito-ing, award-winning reviewer of qualitative re- rial review board. AMJ has published many quali-search for AMJ. Over the past two and a half years, tative papers. The coveted AMJ Best Article AwardI have developed a tremendous respect for Bob’s has been won by three qualitative papers—Gersickkeen eye for evaluating qualitative research sub- (1989), Isabella (1990), and Dutton and Duckerichmissions, and great admiration for the painstaking (1991)—and by one paper that combined qualita-advice he provides authors about how to improve tive and quantitative methods: Sutton and Rafaelli,their work. As a world-renowned qualitative author (1988). Despite these successes, most qualitativehimself, Bob is in an excellent position to provide papers, like most quantitative ones, do not succeedobservations about how authors might increase the in being accepted. This situation is not surprisingchances of having their qualitative research ac- for a journal with a 10 percent acceptance rate.cepted for publication at AMJ. However, it seems to me as a reviewer that there In a three-way electronic mail conversation about are certain recurrent issues in qualitative submis-the challenges and opportunities of qualitative re- sions that, if addressed, could improve the pros-search, Bob, Tom Lee, and I all concluded that pects for positive revise and resubmit decisionsmany authors with potentially very interesting data and ultimate acceptance at AMJ. This editorial of-sets don’t seem to know how to analyze them to fers suggestions to enhance the quality of qualita-their full potential. This is perhaps not surprising, tive research submitted to AMJ. The ideas are basedgiven the clear predominance of quantitative meth- on my experiences as a reviewer for AMJ and as aods and statistics courses over qualitative ones, past Research Methods Division chair. I have alsoparticularly in North America, as well as the inher- been a published qualitative researcher for 26 yearsently greater subjectivity involved in designing and and have one AMJ publication (out of two submis-analyzing qualitative research. As such, we encour- sions). Hopefully these comments will encourageaged Bob to provide a bit of a minitutorial— com- outstanding qualitative research in management.plete with reference citations and examples of An important caveat is necessary at the outset:high-quality papers that use particular qualitative “There are probably rules for writing the persua-approaches—in addition to his observations about sive, memorable and publishable qualitative re-qualitative research submitted to AMJ. search article but, rest assured, no one knows what The result is a longer-than-usual “From the Edi- they are” (Van Maanen, 1998: xxv). The followingtors” column, but one that we believe is well worth comments seek to inspire and inform readers butthe extra reading time for anyone interested in pro- do not specify formulae, algorithms, or criteria forducing, reviewing, or attempting to coax greater producing good qualitative research. Instead, theinsights from qualitative research. We are fortunate column reviews the nature of qualitative research,to have someone with Bob’s expertise share his notes important linkages between theories andobservations, and we hope that his thoughts will methods, reviews key qualitative methodologies,prove useful to researchers for many years to come. and highlights challenges and opportunities in sub- mitting qualitative research to AMJ. Along the way, Sara Rynes helpful examples of qualitative research are cited Incoming Editor and useful resources are noted. These suggestions I am thankful to Sara for inviting me to write this may help authors strengthen the foundations ofeditorial column encouraging scholars to submit their qualitative manuscript submissions.their qualitative research to the Academy of Man- What Is Qualitative Research and Why Is It Important? I wish to thank Tom Lee and Sara Rynes for theirhelpful comments and encouragement in preparing this Qualitative research is multimethod researcheditorial. that uses an interpretive, naturalistic approach to 454
2004 Gephart 455its subject matter (Denzin & Lincoln, 1994). Quali- cepts. Quantitative research codes, counts, and quan-tative research emphasizes qualities of entities— tifies phenomena in its effort to meaningfully repre-the processes and meanings that occur naturally sent concepts. Qualitative research thus has an(Denzin & Lincoln, 2000: 8). Qualitative research inherently literary and humanistic focus, whereasoften studies phenomena in the environments in quantitative research is grounded in mathematicalwhich they naturally occur and uses social actors’ and statistical knowledge. An important value ofmeanings to understand the phenomena (Denzin & qualitative research is description and understandingLincoln, 1994: 2). Qualitative research addresses of the actual human interactions, meanings, and pro-questions about how social experience is created cesses that constitute real-life organizational settings.and given meaning and produces representations of The depiction and understanding of the meanings ofthe world that make the world visible (Denzin & organization members is important in itself (Nelkin &Lincoln, 2000: 3). Beyond this, qualitative research Brown, 1984) and is a task often neglected in organi-is “particularly difficult to pin down” because of its zational research. The domain of naturally occurring“flexibility and emergent character” (Van Maanen, meanings is highly accessible to qualitative research1998: xi). Qualitative research is often designed at and distant from quantitative research. An importantthe same time it is being done; it requires “highly issue is to balance the humanistic and literary aspectscontextualized individual judgements” (Van Maanen, of qualitative research that focus on meanings with1998: xi); morever, it is open to unanticipated the demands for scientific knowledge based in math-events, and it offers holistic depictions of realities ematical or statistical reasoning.that cannot be reduced to a few variables. A second important point is that qualitative re- Clarity can be gained by contrasting qualitative search involves both data collection and data anal-research with quantitative research that “empha- ysis. Both steps in the research process can besizes measurement and analysis of causal relations qualitative or quantitative. Many scholars consideramong variables” (Denzin & Lincoln, 2000: 8). Al- the quantitative analysis of qualitative data to bethough the two research genres overlap, qualitative qualitative research. But it can be argued that quan-research can be conceived of as inductive and in- titative analysis of qualitative data requires data toterpretive (Van Maanen, 1998). It provides a narra- be quantified, and hence this is quantitative re-tive of people’s view(s) of reality and it relies on search. My point is that management researcherswords and talk to create texts. Qualitative work is face many mathematical, statistical, and measure-highly descriptive and often recounts who said ment challenges when they apply quantitative orwhat to whom as well as how, when, and why. An calculative techniques or perspectives to qualita-emphasis on situational details unfolding over time tive data. These challenges become obscured whenallows qualitative research to describe processes. research that uses quantitative tools of analysis isQualitative researchers also seek to explain re- labeled qualitative research.search observations by providing well-substanti- Qualitative research is important for manage-ated conceptual insights that reveal how broad con- ment scholarship for many reasons. In brief, it pro-cepts and theories operate in particular cases. This vides insights that are difficult to produce withapproach is distinct from that of quantitative re- quantitative research. For example, qualitative re-search using the hypothetical-deductive model that search can provide thick, detailed descriptions ofuncovers important relationships among variables actual actions in real-life contexts that recover andand tests general propositions. preserve the actual meanings that actors ascribe to The distinction just drawn between qualitative these actions and settings. Qualitative research canand quantitative research overstates the differences thus provide bases for understanding social pro-between these overlapping genres. But it does call cesses that underlie management. Qualitative re-attention to two critical issues. First, qualitative search can also provide memorable examples ofresearch employs the meanings in use by societal important management issues and concepts thatmembers to explain how they directly experience enrich the field. Finally, qualitative research haseveryday life realities. It builds social science con- potential to rehumanize research and theory bystructs from members’ “concepts-in-use” and fo- highlighting the human interactions and meaningscuses on the socially constructed nature of reality that underlie phenomena and relationships among(Schutz, 1973). Quantitative, positivist research, in variables that are often addressed in the field.contrast, imposes scientific meanings on membersto explain a singular, presumed-to-be true reality The Methodological Importance of Theorythat nonscientists may not appreciate. Second,qualitative research starts from and returns to words, The relationship between theory and methodol-talk, and texts as meaningful representations of con- ogy is important. Researchers need to use method-
456 Academy of Management Journal Augustologies that are consistent with the assumptions grounded theorizing, have been used. Indeed, mostand aims of the theoretical view being expressed. A authors making qualitative submissions claim tosimplified conception of three perspectives used in have used grounded theory processes, although ref-management research is presented in Table 1. Pos- erences to grounded theory are more common thanitivism and postpositivism adopt the stance of real- detailed application of grounded theory tech-ism and rely on the assumption of an objective niques. The problem is that grounded theory oftenworld external to the mind that is mirrored by does not fit well with the objectives of positivist orscientific data and theories. Positivism and post- postpositivist qualitative research. The misfit oc-positivism are efforts to uncover truth or true real- curs in part because, like many other qualitativeity. Postpositivism, the more recent view, differs techniques discussed below, grounded theory orig-from positivism in holding that reality can be inated within the interpretive research tradition ofknown only probabilistically, and hence verifica- social research (Van Maanen, 1998) and was de-tion is not possible. Falsification, not verification, signed to achieve interpretive research goals andof hypotheses becomes the basic task of research. insights concerning meanings, as noted below. ThisWell-developed postpositivist qualitative methods theoretical-methodological inconsistency may incan uncover facts and compare facts to hypotheses part explain why many qualitative research sub-or prior findings in an attempt to falsify prior hy- missions, particularly those in the positivist tradi-potheses or to contradict previous knowledge. tion, provide insights that are somewhat limited A large proportion of the qualitative research I and at times superficial. It is difficult to providehave reviewed for AMJ can be characterized as rep- strong and rigorous findings without well-devel-resenting positivism and postpositivism. Many of oped criteria for evaluating hypotheses. And super-these submissions seek to mirror quantitative re- ficial findings seem likely if grounded theory issearch techniques. An important challenge for this applied in ways that omit analysis of the differ-qualitative research is to articulate rules or bases ences in meanings across important social groups.for deciding “associations” and for determining Two exemplars of positivist research published inhow results and findings fit with preliminary prop- AMJ are McNamara and Bromiley’s (1997) study ofositions or hypotheses. This is a challenge, since decision making using qualitative and quantitativequalitative research lacks the explicit coefficients data, and Gersick’s (1989) discovery-oriented qual-and criteria for evaluating and falsifying hypothe- itative study of groups.ses that quantitative research has developed. The focus of the interpretive perspective differs Perhaps because of this challenge, well-known from the focus on variables and hypothesis falsifi-qualitative methods from social science, such as cation used in postpositivism. The goal of interpre- TABLE 1 Research Traditionsa Positivism and Tradition Postpositivism Interpretive Research Critical PostmodernismAssumptions about reality Realism: Objective reality Relativism: Local intersubjective Historical realism: Material/symbolic that can be understood realities composed from reality shaped by values and by mirror of science: subjective and objective crystallizes over time definitive/probabilistic meanings: represented with concepts of actorsGoal Discover truth Describe meanings, Uncover hidden interests and understanding contradictions: critique, transformation, and emancipationTasks Undertake explanation Produce descriptions of Develop structural or historical and control of members’ meanings and insights that reveal contradictions variables: discern definitions of situation: and allow emancipation, spaces verified hypotheses or understand reality for silenced voices nonfalsified hypotheses constructionUnit of analysis Variable Verbal or nonverbal action Contradictions, critical incidents, signs and symbolsMethods focus Uncover facts, compare Recover and understand Understand historical evolution of these to hypotheses or situated meanings, systematic meanings, material practices, propositions divergences in meaning contradictions, inequalities a This table is based on Gephart (1999), Guba and Lincoln (1994), and Lincoln and Guba (2000).
2004 Gephart 457tive research is to understand the actual production tive methodologies to uncover divergent meaningsof meanings and concepts used by social actors in held by groups in power-laden relationships. Mor-real settings. A relativist stance is adopted such row (1994) provides a helpful discussion of criticalthat diverse meanings are assumed to exist and to theory methodology. Given the theoretical focus ofinfluence how people understand and respond to critical research, many critical management papersthe objective world. Interpretive research thus de- have appeared in the Academy of Management Re-scribes how different meanings held by different view. But empirical research that uses critical the-persons or groups produce and sustain a sense of ory is rare in management (Alvesson & Wilmott,truth, particularly in the face of competing defini- 1992) and would be welcome at AMJ (Eden, 2003).tions of reality. And it inductively constructs social Ashcraft (2001) offers an example of critical femi-science concepts using concepts of social actors as nist research in AMJ.the foundations for analytic induction. This con- Like critical research, postmodern thought alsocern with meanings and second-order concepts— begins with the assumption that realities are valuethe concepts of the concepts of social actors—leads laden and contain contradictions. But postmodernto a focus on thick descriptions of members’ talk thought tends to focus on signs and symbols andand nonverbal actions in specific settings. Rather the idea that these are decoupled from realities theythan producing qualitative facts to evaluate hy- represent. As such, postmodern thought adds a fo-potheses, interpretive researchers seek to describe cus on texts or written documents that symboli-and understand members’ meanings and the impli- cally create and disclose structured inequalities.cations that divergent meanings hold for social in- Critical postmodern thought has thus begun to uti-teraction. Isabella’s (1990) award-winning paper lize textual, literary, and deconstructionist ap-stands as an excellent example of interpretive re- proaches to analysis of materials. Boje’s (1995)search published in AMJ. study of multiple discourses at Disney provides an Critical postmodernism combines critical theory example of postmodern research with a critical fla-and postmodern thought. Critical research de- vor that appeared in AMJ.scribes the historical emergence of social structures This brief review of theoretical perspectives il-and the contemporary contexts in which these lustrates three distinctive approaches to theory thatstructures form contradictions with implications are related to research methodology. Postpositivismfor social action and human freedom. For example, requires methods of collecting and analyzing fac-critical research explores the presence and impli- tual depictions of the world that reveal singularcations of the basic contradiction of advanced cap- truths or realities and that can be used to evaluateitalism: the desire for profit exceeds the available (falsify) hypotheses. Interpretive research uncov-profit. Contradictions are conceived to be basic to ers, describes, and theoretically interprets actualthe exploitation that emerges when hegemonic meanings that people use in real settings. It exam-worldviews conceal contradictions, leaving people ines how particular meanings become shared, dom-unaware of tacit forms of domination and subjuga- inant, and/or contested in situations in which al-tion that are present. Critical research uncovers ternative meanings and understandings are presentrelations of dominance and subjugation and pro- and possible. Critical postmodernism describesduces insights to make social actors reflexively dominant and subordinated meanings, displays theaware of their own role in the reproduction of cap- power implications of meanings, and encouragesitalist inequities. Critical research seeks to trans- critical reflexivity to make people aware of theform the social order and allow emancipation from constraints on their own meanings and actions.unwanted structures of domination. Critical reflexivity provides a means for emancipa- Methodologically, critical research emphasizes tion from structures of domination.dialogic and dialectical methods (Lincoln & Guba, Clearly, qualitative methodologies must be used2000) as ways to transcend taken-for-granted in ways that are consistent with the theoretical ortruths. Critical research adopts a historical realist paradigmatic view(s) adopted and the specificassumption that the construction of reality is problems being explored. This consistency is im-shaped by social, political and economic values portant so that the research process is capable ofthat crystallize and become reified over time. This producing the kinds of data and analyses necessi-constructed reality is experienced as firmly as if it tated by the theory in use and the goals of researchwere the unconstructed reality assumed by positiv- in the related paradigm. Two options could en-ists. Thus, critical research uncovers facts about hance consistency in theories and methodologies.power relations that are obscure to societal mem- First, scholars could adopt postpositivist method-bers. Further, its assumption is that there are mul- ological techniques from social science to enhancetiple views of the world, and it employs interpre- consistency between postpositivist theory and
458 Academy of Management Journal Augustmethods-in-use in management. Second, scholars gest resources to consult when planning qualitativecould use interpretive or critical postmodern per- research, and note recent AMJ papers that use thesespectives more often and adopt social science methodologies.methods that were originally developed for inter- A case study is research that describes a singlepretive and critical research agendas and purposes. event or unit of analysis determined by the re-Most AMJ authors and reviewers are well skilled searcher. There are different types of case studiesand trained in quantitative, positivist techniques (Hamel, Dufour, & Fortin, 1993). Case studies oftenand perspectives but are less prepared to produce use archival or documentary data along with otherinterpretive and critical postmodern research. This sources, combine qualitative and quantitative data,discrepancy may explain why interpretive and crit- and examine a phenomenon or “case” as it changesical postmodern research is less common in AMJ over time. A well-known example of case studythan positivist research. However, I believe and research is Biggart’s (1977) classic study of changehave been assured by Tom Lee and Sara Rynes that at the U.S. post office. Another example is Hera-AMJ values and welcomes submissions from each cleous and Barrett’s (2001) nicely done case studyof these three perspectives. of the implementation of electronic trading on the London Insurance Market, which was published in AMJ.Well-Developed Methodologies Are Useful Interviews are situated, face-to-face interactions Qualitative research requires qualitative methods in which researchers typically pose questions thatby definition. It is important to show what was respondents answer. There are different types ofdone in the research process and to articulate how interviews and related methodologies. Ethno-research practices transformed observations into graphic interviews (Spradley, 1979) are used to un-data, results, findings, and insights. The methodol- derstand informants’ conceptions of culture. Longogy used need not be complex, and the method- interviews (McCracken, 1988) link analytical cate-ological account need not dominate the written gories and literature with respondents’ cultural cat-report. But many qualitative submissions I have egories and meanings. Focus groups assemblereviewed lacked explicit analytical methods. groups of individuals who respond to questions or The major problem with failure to use a rigorous, themes. They represent a collective rather than in-well-developed methodology is that data are un- dividualistic research method that permits collec-likely to be systematically, comprehensively, or ex- tive testimonies and narratives (Madriz, 2000: 836).haustively reviewed. Hence, findings produced A classic interview-based study in AMJ is Isabella’sfrom informal or ill-defined procedures may be (1990) paper on organizational change.both different from and weaker than those pro- A number of observational methods are availableduced when a clear methodological process is for use. The first method is participant observation,used. When methods are used but not described which involves social interaction in the field withexplicitly, or when findings are presented early in a subjects, direct observation of relevant events, for-study and prior to discussion of goals, theory, and mal and informal interviewing, some counting, col-methods, other problems arise. For example, if it is lection of documents, and flexibility in the direc-unclear to the reader how research was undertaken, tion the study takes (McCall & Simmons, 1969: 1).it may be difficult to connect claims in the paper In participant observation, it is common for a re-that reports that research to the data presented. The searcher to play the role of a member of the groupoperation of concepts in data needs to be revealed studied and to use subjective experiences as criticalin clear and explicit ways if the findings are to be data. Barker’s (1993) study of how teams controlcomprehensible and credible. While qualitative members’ behavior provides a classic example ofmethods need to be elaborated or modified for each observation-based research. Yakura (2002) pro-new application, this does not mean that anything vides a recent example of participant-observation-goes or that the best method is no method. Re- based research published in AMJ. A second obser-searchers need to report their sources and types of vational approach is ethnography (Hammersley &data as well as their data analysis practices. Atkinson, 1995), which involves the production of Qualitative data are collected using one or more descriptions of culture obtained by immersion inresearch approaches, including case studies, inter- the culture studied. Perlow, Okhuysen, and Repen-views, observations, grounded theory, and textual ning, (2002) provide a recent example of ethno-analysis. General overviews of qualitative research graphically informed fieldwork published in AMJ.may be found in Silverman (2004) and Golden- A third observation-based approach is ethnometh-Biddle and Locke (1997). In this section, I provide odology (Coulon, 1995), defined as the study of thea brief overview of these useful methodologies, sug- practical methods members of society use to con-
2004 Gephart 459struct and maintain a sensible understanding of the of management texts is an example of narrative-social world. An example of ethnomethodological rhetorical analysis published in AMJ.research published in AMJ is provided by my paper Textual analysis can also be undertaken withon disaster sensemaking (Gephart, 1993). computer software support (Kabanoff, 1997). Com- Two additional observational methods have im- puter-aided textual analysis uses the capabilities ofportant but unrealized potential in management computers to produce qualitative and numericalresearch. The first is conversational analysis, the results from qualitative or textual materials (Kelle,study of sequential, utterance-by-utterance, talk 1995). Computer-aided interpretive textual analysisand conversation that often uses ethnomethod- is a related qualitative research approach that pro-ological concepts to provide an understanding of vides insights into organization members’ mean-how talk structures social interaction (Gubrium & ings by using computers to support theoreticalHolstein, 2000: 492). The second is systematic self- sampling, textual analysis, expansion analysis,observation, a new and well-developed observa- and grounded theory development (Gephart, 1997).tional technique that involves “training informants Computer-supported qualitative data analysis al-to observe and record a selected feature of their lows one to systematically, comprehensively, andown everyday experience” (Rodriguez & Ryave, exhaustively analyze a corpus of data. Many qual-2002: 2). Systematic self-observation may prove itative papers submitted to AMJ, particularly posi-particularly useful to researchers interested in lan- tivism-oriented papers, would benefit from a com-guage use in organizations. puter-supported textual analysis approach because Grounded theorizing (Glaser & Strauss, 1967) is it provides ways to investigate qualitative andthe process of iteratively and inductively construct- quantitative features of texts and offers approachesing theory from observations using a process of to hypothesis testing using qualitative and/or quan-theoretical sampling in which emergent insights titative data. Few AMJ papers have used such tech- niques even when these have been recommendeddirect selection and inclusion of the “next” infor- during the review process.mant or slice of data. Grounded theory involvesconstant comparative analysis whereby groups arecompared on the basis of theoretical similarities Challenges and Opportunitiesand differences. A large number of research sub- This section outlines common problems andmissions and qualitative papers published in AMJ challenges found in many qualitative submissionsrefer to grounded theory as part of their methodol- to AMJ, and potential solutions to these problems.ogy. Indeed, by examining the methodological These points follow from the issues raised above, ascitations in qualitative submissions, one would well as from rereading reviews written by otherconclude grounded theory was a ubiquitous meth- reviewers and myself in the last two years, andodology in our field. But relatively few manuscripts editors’ letters to authors in which these materialsexplain how grounded theory methodology was had been retained.used to produce results and findings. It is even less The first issue is that many submissions appearcommon for qualitative papers to address related to be “one off” papers that do not seem to be em-grounded theory practices, such as theoretical bedded in ongoing research projects or programs.sampling and the constant comparative method of Qualitative research manuscripts that emerge fromanalysis. Perlow, Okhuysen, and Repenning (2002) broad, ongoing research programs seem more likelyprovide a recent example of grounded theory– to produce substantial new insights because theyinfluenced research published recently in AMJ. address multiple issues and have large corpora of Textual analysis involves analysis of texts using data to analyze. This point is underscored whenideas from theories in hermeneutics and literary authors revise and resubmit a paper. Since fewcriticism intended to provide systematic under- manuscripts are acceptable on first submission, re-standing of texts. Two forms of textual analysis that viewers often request additional data and analyses.have been used in management and organizational But few authors actually return to the field, collectresearch are semiotics, which is the study of signs new data or add previously collected data, or em-(Barley, 1983), and narrative analysis (Boje, 2001), ploy new or different analytical procedures. Wherewhich examines structural, literary features of research is part of an ongoing research program,texts. Rhetorical analysis of texts (Simons, 1989) is authors can more readily elaborate their ideas,also relevant to narrative analysis. Narrative-rhetor- modify their topics, and analyze additional data.ical analysis is illustrated by Barry and Elmes’s The iterative nature of qualitative research should(1997) analysis of strategic management as a form continue during the submission and the review andof fiction. Locke and Golden-Biddle’s (1997) study revision stages of research.
460 Academy of Management Journal August A second problem is that the introductions to tions: to clearly describe the processes used to re-qualitative papers often lack adequate reviews of view data and to formulate themes and insights.important literature relevant to the topics of the The reader needs to know how categories or themespapers. A surprising number of qualitative papers were discerned in data and how key decisions wereprovide literature reviews as part of their results, made in the research process. It is useful to refer tofindings, or conclusions and only after results and explicit and established research methods and lit-findings have been stated. This practice makes the erature to describe general methodological ap-work completely mysterious until topics, concepts, proaches and to indicate how such methods haveand past research are finally noted. By the time this been modified or adapted to address current re-occurs, the findings often appear to readers to have search questions and data. But methodology shouldbeen arbitrarily assembled or drawn directly from be explained and then used. It should not over-the literature rather than based on data, causing whelm the conceptual importance of a paper.reviewers to ask, What is new here? This problem Several specific methodological issues are oftencan be addressed by providing an effective review evident to reviewers once data are presented. Aof literature that notes the content and limits of common reviewer request is to provide the “thickerprior research in the apposite field and that points and more detailed” descriptions that are essentialto a lacuna in the literature that the study can for capturing members’ meanings and in situ socialaddress. Further, qualitative papers need to address processes. Thus it is important where possible toimportant research in related fields as well as in include raw or primary qualitative data in papersmanagement since management is a transdisci- (for instance, actual talk by respondents). It is alsoplinary field and significant implications are often important to analyze or interpret such data, notbased in or relevant to important issues and social simply to present it. In addition, it is important toresearch trends outside the field. compare and contrast examples to reveal concep- A third and related problem is that qualitative tual similarities and differences in data. Thesesubmissions often fail to state explicit goals, objec- examples need to represent key concepts and totives, or research questions that frame the papers be selected on conceptual and methodologicaland guide data analysis and research outcomes. It is grounds, with discussion provided as to how theimportant for qualitative research to have a clear examples relate to the broader corpus of data usedfocus and bases on which to proceed. Also, the in the study. Drawing these links avoids the com-importance of the research questions posed is fun- mon problem of “exampling,” whereby a researcherdamental to the contribution made by a given pa- addresses a few examples but fails to explain howper. Through specification of research questions these examples represent a broader data set or tothat reflect an important gap in the literature, a explain why they were chosen. Finally, there is astudy can identify important lacunae in the schol- tendency for qualitative submissions to presentarly domain. faits accomplis, offering findings without explana- Fourth, where questions are provided, the con- tion as to their origins. This practice is a problemcepts underlying them often are not well defined, since it is important to show how findings wereand the meaning of the questions remains elusive. surfaced from data or otherwise disclosed throughIt is important for research papers—whether qual- analysis. Without these connections, findings oftenitative or quantitative—to define and explain key appear to lack grounding in data.concepts in ways that allow the reader to anticipate A sixth domain of problems concerns discussionhow the concepts could be located in data or ob- and conclusion sections. Authors need to revisitservations. Conceptual and empirical definition of research questions or goals in their discussions tokey concepts is important even when a paper’s explain how their questions were answered andauthors seek to dispute or elaborate prior defini- how their goals were achieved in the reported re-tions. And the theoretical background to these con- search. The broader implications and importancecepts needs to be disclosed in ways that create of the findings are contributions the paper offers.consistency among theories, concepts, research These need to be explained and related to issues inquestions, and methodologies. management and to key social science research Fifth, although methodological issues are impor- issues.tant to qualitative research, it is extremely commonto find that the methodology is underspecified. ConclusionSince methodological issues have already been ad-dressed in detail, only a few brief comments are Good qualitative research is difficult and chal-noted here. It is important to describe the analytical lenging to undertake. Many scholars believe goodmethod or approach used to address research ques- qualitative research is more difficult and time con-
2004 Gephart 461suming to create than good quantitative research. N. K. Denzin & Y. S. Lincoln (Eds.), Handbook ofQualitative research often involves fieldwork, and qualitative research (2nd ed.): 1–28. Thousandthe word “work” is important here. There are no Oaks, CA: Sage.algorithms for producing it. Qualitative researchers Dutton, J. E., & Duckerich, J. M. 1991. Keeping an eye onwill likely be less productive than quantitative re- the mirror: Image and identity in organizational ad-searchers in terms of the number of manuscripts aptation. Academy of Management Journal, 34:produced. Qualitative researchers should be evalu- 517–554.ated in terms of the significance and the impact Eden, D. 2003. Critical management studies and thetheir publications have on the field. The advantage Academy of Management Journal: Challenge andof qualitative research is that it offers scholars a counterchallenge. Academy of Management Jour-rewarding and meaningful way to lead their lives. nal, 46: 390 –394.The rewards include direct engagement with every- Gephart, R. P. 1993. The textual approach: Risk andday management and organizational realities and blame in disaster sensemaking. Academy of Man-opportunities to make substantial contributions to agement Journal, 36: 1465–1514.the field. Qualitative research often advances the Gephart, R. P. 1997. Hazardous measures: An interpre-field by providing unique, memorable, socially im- tive textual analysis of quantitative sensemakingportant and theoretically meaningful contributions during crises. Journal of Organizational Behavior,to scholarly discourse and organizational life. 18: 583– 622. Robert P. Gephart, Jr. Gephart, R. P. 1999. Paradigms and research methods. University of Alberta Research Methods Division forum, 4. aom.pace.edu/ rmd/1999_RMD_Forum_Paradigms_and_Research_ Methods.htm. REFERENCES Gersick, C. 1989. Marking time: Predictable transitions in task groups. Academy of Management Journal, 32:Alvesson, M., & Willmott, H. 1992. Critical management 274 –309. studies. Thousand Oaks, CA: Sage. Glaser, B., & Strauss, A. 1967. The discovery ofAshcraft, K. L. 2001. Organized dissonance: Feminist grounded theory. Chicago: Aldine Press. bureaucracy as hybrid form. Academy of Manage- ment Journal, 44: 1301–1322. Golden-Biddle, K., & Locke, K. 1997. Composing quali-Barker, J. R. 1993. Tightening the iron cage: Concertive tative research. Thousand Oaks, CA: Sage. control in self-managing teams. Administrative Sci- Guba, E. G., & Lincoln, Y. S. 1994. Competing paradigms ence Quarterly, 38: 408 – 437. in qualitative research. In N. K. Denzin & Y. S. Lin-Barley, S. 1983. Semiotics and the study of organiza- coln (Eds.), Handbook of qualitative research (2nd tional and occupational cultures. Administrative ed.): 105–117. Thousand Oaks, CA: Sage. Science Quarterly, 28: 393– 413. Gubrium, J. F., & Holstein, J. A. 2000. Analyzing inter-Barry, D., & Elmes, M. 1997. Strategy retold: Toward a pretive practice. In N. K. Denzin & Y. S. Lincoln narrative view of strategic discourse. Academy of (Eds.), Handbook of qualitative research (2nd ed.): Management Review, 22: 429 – 452. 487–508. Thousand Oaks, CA: Sage.Biggart, N. 1977. The creative-destructive process of or- Hamel, J., Dufour, S., & Fortin, D. 1993 Case study meth- ganizational change. Administrative Science Quar- ods. Thousand Oaks, CA: Sage. terly, 22: 410 – 426. Hammersley, M., & Atkinson, P. 1995. Ethnography:Boje, D. 1995. Stories of the storytelling organization: A Principles in practice (2nd ed.). London: Routledge. postmodern analysis of Disney as Tamara-land. Heracleous, L., & Barrett, M. 2001. Organizational change Academy of Management Journal, 38: 997–1035. as discourse: Communicative actions and deepBoje, D. 2001. Narrative methods for organizational structures in the context of information technology and communication research. Thousand Oaks, CA: implementation. Academy of Management Jour- Sage. nal, 44: 755–778.Coulon, A. 1995. Ethnomethodology. Thousand Oaks, Isabella, L. 1990. Evolving interpretation as a change CA: Sage. unfolds: How managers construe key organizationalDenzin, N., & Lincoln, Y. S. 1994. Introduction: Entering events. Academy of Management Journal, 33: the field of qualitative research. In N. K. Denzin & 7– 41. Y. W. Lincoln (Eds.), Handbook of qualitative re- Kabanoff, B. 1997. Introduction—Computers can read as search: 1–17. Thousand Oaks, CA: Sage. well as count: Computer-aided text analysis in or-Denzin, N. K., & Lincoln, Y. S. 2000. Introduction: The ganizational research. Journal of Organizational discipline and practice of qualitative research. In Behavior, 18: 507–511.
462 Academy of Management Journal AugustKelle, U. 1995. Computer-aided qualitative data anal- Perlow, L., Okhuysen, G., & Repenning, N. P. 2002. The ysis. Thousand Oaks, CA: Sage. speed trap: Exploring the relationship between deci-Lincoln, Y. S., & Guba, E. G. 2000. Paradigmatic contro- sion making and temporal context. Academy of versies, contradictions and emerging confluences. In Management Journal, 45: 931–955. N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Rodriguez, N., & Ryave, A. 2002. Systematic self-obser- qualitative research (2nd ed.): 163–188. Thousand vation (Qualitative Research Methods Series, no. Oaks, CA: Sage. 49). Thousand Oaks, CA: Sage.Locke, K., & Golden-Biddle, K. 1997. Constructing oppor- Schutz, A. 1973. Concept and theory formation in the tunities for contribution: Structuring coherence and social sciences. In M. Natanson (Ed.), Alfred Schutz “problematizing” in organization studies. Academy collected papers I: The problem of social reality: of Management Journal, 40: 1023–1062. 48 – 66. The Hague: Martinus Nijhoff.Madriz, E. 2000. Focus groups in feminist research. In Silverman, D. (Ed.). 2004. Qualitative research: Theory, N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of method and practice (2nd ed.). Thousand Oaks, CA: qualitative research (2nd ed.): 835– 850. Thousand Sage. Oaks, CA: Sage. Simons, H. W. 1989. Rhetoric in the human sciences.McCall, G. J., & Simmons, J. L. 1969. The nature of par- Thousand Oaks, CA: Sage. ticipant observation. In G. J. McCall & J. L. Simmons (Eds.), Issues in participant observation: A text and Spradley, J. P. 1979. The ethnographic interview. New reader: 1–5. Reading, MA: Addison-Wesley. York: Holt, Rinehart & Winston.McCracken, G. 1988. The long interview (Qualitative Sutton, R. I., & Rafaeli, A. 1988. Untangling the relation- Research Methods Series, no. 13). Thousand Oaks, ship between displayed emotions and organizational CA: Sage. sales: The case of convenience stores. Academy ofMcNamara, G., & Bromiley, P. 1997. Decision making in an Management Journal, 31: 461– 487. organizational setting: Cognitive and organizational in- Van Maanen, J. 1998. Different strokes: Qualitative re- fluences on risk assessment in commercial lending. search in the Administrative Science Quarterly from Academy of Management Journal, 40: 1063–1088. 1956 to 1996. In J. Van Maanen (Ed.), QualitativeMorrow, R. 1994. Critical theory and methodology. studies of organizations: ix–xxxii. Thousand Oaks, Thousand Oaks, CA: Sage. CA: Sage.Nelkin, D., & Brown, M. S. 1984. Workers at risk: Voices Yakura, E. 2002. Charting time: Timelines as temporal from the workplace. Chicago: University of Chicago boundary objects. Academy of Management Jour- Press. nal, 45: 956 –970. PAST EDITORS Paul M. Dauten, Jr. University of Illinois 1958 – 60 Dalton E. McFarland Michigan State University 1961– 63 Paul J. Gordon Indiana University 1964 – 66 Stanley C. Vance University of Oregon 1967– 69 William G. Scott University of Washington 1970 –72 John B. Miner Georgia State University 1973–75 Larry L. Cummings University of Wisconsin–Madison 1976 –78 John W. Slocum, Jr. Southern Methodist University 1979 – 81 Thomas A. Mahoney Vanderbilt University 1982– 84 Janice M. Beyer New York University 1985– 87 Richard T. Mowday University of Oregon 1988 –90 Michael A. Hitt Texas A&M University 1991–93 Angelo S. DeNisi Rutgers University 1994 –96 Anne S. Tsui Hong Kong University of Science & Technology 1997–99 Gregory B. Northcraft University of Illinois at Urbana-Champaign 2000 – 01