Introduction: Eclecticism in Methods —David A. Harrison
Controlling Method Effects in Self Report Instruments —Mary E. McLaughlin
Missing Data: Instrument-Level Heffalumps and Item-Level Woozles —Philip L. Roth and Fred S. Switzer III
Paradigms and Research Methods —Robert Gephart
Improving the Power of Moderated Multiple Regression to Estimate Interaction Effects —Herman Aguinis and Charles A. Pierce
Paradigms and Research Methods
University of Alberta
There has been considerable interest in recent years in the role of philosophical assumptions and paradigms in doing research. During the 1970's and 1980's prominent concerns were raised about the limits of quantitative data and methods often associated with positivism, the prevailing paradigm. Positivism assumes an objective world which scientific methods can more or less readily represent and measure, and it seeks to predict and explain causal relations among key variables. Critics argued that positivistic methods strip contexts from meanings in the process of developing quantified measures of phenomena (Guba and Lincoln, 1994: 106). In particular, quantitative measures often exclude members' meanings and interpretations from data which are collected. These methods impose outsiders meanings and interpretations on data. And they require statistical samples which often do not represent specific social groups and which do not allow generalization to or understanding of individual cases. Finally, quantitative and positivistic methods tend to exclude discovery from the domain of scientific inquiry.
Positivism has become a dominant institutional form in social research. Yet this dominance is increasingly challenged by critics from two alternative traditions -- interpretive constructionism and critical postmodernism -- which are well established and which have played prominent roles in Western thought. Constructionism and critical postmodernism raise fundamental philosophical challenges for positivism and offer alternative theoretical, methodological and practical approaches to research on management and organizations. These traditions have garnered enhanced interest in part because they address timely social and political issues which positivist research has tended not to address.
Positivistic concerns to uncover truths and facts using experimental or survey methods have been challenged by interpretivists who assert that these methods impose a view of the world on subjects rather than capturing, describing and understanding these world views. Critical postmodernists argue that these imposed views or measures also implicitly support forms of scientific knowledge that explicitly reproduce capitalist structures and associated hierarchies of inequality.
This brief paper seeks to examine the three paradigms or worldviews which are prominent in contemporary social research -- positivism, interpretivism and critical postmodernism. Interpretivism and critical postmodernism are increasingly common in management and organizational scholarship though they are not so well understood. The paper seeks to summarize the key features of each worldview, the nature of knowledge pursued, and the different means by which knowledge is produced and assessed within each paradigm or worldview. The intent is to help readers understand some of the basic assumptions underlying forms of research present in the field. Further, the paper will hopefully sensitize the reader to the methodological issues which emerge in interplay between theories and forms of knowledge, social practices, and research methods. Better understanding of methodological issues may encourage improved research practices by fostering consistency between the underlying assumptions, theories and knowledge production activities of management and organizational researchers. Further, since many actual research studies use aspects of more than one paradigm, the discussion here is not intended to advocate one paradigm as opposed to another. Rather, distinctions offered here are intended to clarify different strands of thinking and researching so that the strands can be interwoven into the research texts which emerge as consistent and coherent, true to their underlying premises even as they advance these premises and combine insights into new and interesting ways.
A final point of introduction relates to the history of this paper. The paper is based on a typology of paradigms and research methods originally presented to a Research Methods Division Workshop presented to the Western Academy of Management (Gephart, 1995), and presented occasionally thereafter at other workshops at Academy. Table 1 summarizes the key paradigms and the features of the paradigms discussed below. In addition, the paper is very directly informed by ideas in Guba and Lincoln (1996). And the usefulness of the paradigms schema as a pedagogical tool was substantiated in 3 successive offerings of Organizational Analysis 701, the Ph.D. Seminar on Organization Theory I have offered at the University of Alberta. Readers interested in further information on the workshops and Ph.D. seminar are invited to contact me directly since the paper which follows is at best a very brief overview, not a complete introduction to the domains. I hope it is helpful!
MANAGEMENT RESEARCH PARADIGMS
|ASSUMPTIONS||Objective world which science can 'mirror' with privileged knowledge||Intersubjective world which science can represent with concepts of concepts of actors; social construction of reality||Material world of structured contradictions and/or exploitation which can be objectively known only by removing tacit ideological biases|
|KEY FOCUS or IDEAS||Search for contextual and organizational variables which cause organizational actions||Search for patterns of meaning||Search for disguised contradictions hidden by ideology; open spaces for previously silenced voices|
|KEY THEORIES IN PARADIGM||Contingency theory; systems theory; population ecology; transaction cost economics of organizing; dustbowl empiricism||Symbolic interaction; ethnomethodology; phenomenology; hermeneutics||Marxism; critical theory;
PM: poststructuralism; postmodernism; deconstructionism; semiotics
|KEY FIGURES||Lorsch and Lawrence; Hannan and Freeman; Oliver Williamson||Goffman; Garfinkel, Schutz; Van Maanen, David Silverman||Marx; Habermas: Offe|
|GOAL OF PARADIGM||Uncover truth and facts as quantitatively specified relations among variables||Describe meanings, understand members' definitions of the situation, examine how objective realities are produced||Uncover hidden interests; expose contractions; enable more informed consciousness; displace ideology with scientific insights; change|
|NATURE OF KNOWLEDGE or FORM OF THEORY||Verified hypotheses involving valid, reliable and precisely measured variables||Abstract descriptions of meanings and members= definitions of situations produced in natural contexts||Structural or historical insights revealing contradictions|
|CRITERIA FOR ASSESSING RESEARCH||Prediction=Explanation
Rigor; internal & external validity, reliability
Basis for action, change potential and mobilization
|UNIT OF ANALYSIS||The variable||Meaning; symbolic act||Contradictions, incidents
PM: the sign
TYPE(S) OF ANALYSIS
questionnaires; secondary data analysis; quantitatively coded documents
Quantitative: regression; Likert scaling; structural equation modeling
Qualitative: grounded theory testing
observation; interviews; conversational analysis; grounded theory
Case studies; conversational and textual analysis; expansion analysis
|Field research, historical
analysis, dialectical analysis
PM: deconstruction, textual analysis
POSITIVISM AND POSTPOSITIVISM
Positivism assumes an objective world hence it often searches for facts conceived in terms of specified correlations and associations among variables. Post positivism is a recent evolution of positivism. Post positivism is consistent with positivism in assuming that an objective world exists but it assumes the world might not be readily apprehended and that variable relations or facts might be only probabilistic, not deterministic. Thus the positivist focus on experimental and quantitative methods used to test and verify hypotheses have been superceded or complemented to some extent by an interest in using qualitative methods to gather broader information outside of readily measured variables. Logically, (i.e. in principle if not in practice) there is a focus on falsification rather than verification given the complexity of real world phenomena -- only one counter-example or feature is needed to falsify a proposed relationship but one must assess all possible variables to verify a relationship is consistent across all such conditions. Further, increasing effort is devoted to establishing the domain of generalizability of findings based on the features of the sample and sampling context.
The recent focus in post positivism has been on qualitative methods modeled on positivistic methods and experimental designs (e.g. Miles and Huberman, 1993). This reflects the attempts by post positivists to address the methodological challenges to quantitative methods which were noted in the introduction to this paper. Increasingly, grounded theory is being used to post-positivists to examine and assess variables and their relationships in situations where quantitative measurement and statistical controls are not possible. It is important to note however that the uses of grounded theory in post-positivism are oriented to confirmation and validation or falsification of hypotheses and to uncovering or surfacing relationships among variables, in contrast to uses of grounded theory in interpretive research where it is applied to understand important distinctions and patterns in members' meanings. One can expect that in future, post positivists will remain concerned to develop methods which preserve contexts and broader meanings associated with data and in general to try to develop approaches to deal with the other problems with positivist methods which interpretivists and critical postmodernists have identified.
INTERPRETIVISM AND CONSTRUCTIVISM
Interpretive research is fundamentally concerned with meaning and it seeks to understand social members' definition of a situation (Schwandt, 1994: 118). Interpretive theory involves building a second order theory or theory of members' theories (Schutz, 1973) in contrast to positivism which is concerned with objective reality and meanings thought to be independent of people. Interpretivists assume that knowledge and meaning are acts of interpretation hence there is no objective knowledge which is independent of thinking, reasoning humans. Interpretivism often addresses essential features of shared meaning and understanding whereas constructivism extends this concern with knowledge as produced and interpreted to an anti-essentialist level. Constructionists argue that knowledge and truth are the result of perspective (Schwandt, 1994: 125) hence all truths are relative to some meaning context or perspective.
There are many interpretivist and constructionist genres but central to all of these has been a concern with subjective meanings -- how individuals or members of society apprehend, understand and make sense of social events and settings (the idea of interpretation) and how this sensemaking produces features of the very settings to which sensemaking is responsive (the concern for reflexivity). Constructionists have also been particularly concerned with the interplay of subjective, objective and intersubjective knowledge. Intersubjectivity is the process of knowing others' minds and the question of intersubjectivity -- how we know others minds -- has been a longstanding challenge (scandal!) in philosophy (Schutz, 1973). Intersubjectivity occurs through language, social interaction, and written texts. A key form of interpretive research is social constructionism (Berger and Luckmann, 1967; Knorr-Cetina, 1981; Gephart, 1978) which seeks to understand the social construction dialectic involving objective, intersubjective and subjective knowledge. This research investigates how the objective features of society (e.g. organizations, social classes, technology and scientific facts) emerge from, depend on, and are constituted by subjective meanings of individuals and intersubjective processes such as discourses or discussions in groups (c.f. Gephart, 1993, 1997). In a sense, interpretivist constructivism "brackets" objective reality and seeks to show how variations in human meanings and sensemaking generate and reflect differences in reified or objective realities.
Given the concern with understanding members' meanings, interpretive researchers have often preferred meaning (versus measurement) oriented methods. In particular, data collection and representation have been accomplished with informant interviewing (Spradley, 1979), ethnography, or the thick description of cultures based on intimate knowledge and participation (Van Maanen, 1988), and even ethnographically linked textual analyses (Gephart, 1993) which use transcripts or verbal protocols of meetings as data. Such verbal or conversational data are collected to represent interactions in important, naturally occurring social settings. And the conversations themselves are seen to constitute important interactions in and features of these settings.
The primary analytic methods used in interpretive research are grounded theorizing and expansion analysis. The basis of grounded theorizing is the idea that positivism is oriented to testing and confirmation of general theories which take the form of well validated propositions which specify relations, often causal, among well defined and quantitatively measured variables. The result is that positivism is effective for theory testing but it precludes theory development and discovery because positivist methods are inherently oriented to testing pre-established hypotheses and propositions. Grounded theory emerged in an era noted for the proliferation of highly abstract grand or "great man" theories which were difficult to empirically assess. That is, social researchers found that there were insufficient linkages between these highly abstract grand theories and the actual issues observed in field research settings. Grounded theory (Glaser and Strauss, 1967) was developed to encourage the development of low level or grounded theories to represent the patterns surfaced in low level or setting specific field research, a mid-range level of theory which was ignored in grand social theorizing. It also sought to attempt to ground or use these highly abstract theories or concepts in specific contexts and research endeavors as a way of determining if the concepts or general theories did have empirical relevance and practical usefulness.
In grounded, qualitative research one often begins with sensitizing or orienting concepts which provide the researcher with a "general sense of reference and guidance in approaching empirical instances" (Blumer, 1969:148.). The sensitizing ideas are examined or applied through micro level observations of interactions in specific social settings. Sensitizing concepts are thereby elaborated and further developed to capture and reflect discovered features of the phenomenon examined. Data are collected often in the form of field notes of interactions and conversations with an interest in surfacing a focus and collecting multiple examples of the phenomenon of emergent interest. Sensitizing concepts act as theoretical lenses to help the researcher find examples as well as patterns in the meanings represented in data, using theoretical sampling rather than random sampling to identify examples of research interest. Constant comparative analysis (Glaser and Strauss, 1967: 22-24 and 101+) provides an alternative to statistical analysis. Essentially this comparative analysis process examines all data slices which are similar on a given dimension or category and compares these to slices which are similar on one or more dimensions but differ on theoretically important dimensions. For example, all statements by workers describing their experiences of risk and danger during an accident can be collected and compared to statements by managers concerning their experiences of risk and danger. This process is used to generate theoretical properties for a category or concept of interest in contrast to other properties and categories. For example, here it could be used to understand different organizational positions (an important, prior categorization of informants) and their experiences of work as involving risk, danger, fear and a sense of safety (emergent theoretical properties of different categories of actors and work practices -- see Gephart, 1993 for further discussion of this example). Constant comparative analysis is completed by 1) comparing all incidents relevant to a given theoretically meaningful category, 2) integrating the categories and their properties, 3) delimiting the range of the theory and then 4) writing the theory (Glaser and Strauss, 1967: 105).
The analytical process often involves expansion analysis of data displays which is similar in many ways to the positivist concern to interpret tables of quantitative results. Expansion analysis (Cicourel, 1980; Gephart, 1993) involves presenting some key example or vignette of data, often qualitative data such as a key story or passage from a conversation, and then providing a discussion of how the conceptual or theoretical processes of interest operate in the data display. The data display is often treated as representative of many similar examples, hence the expansion analysis of one display or example is often generalized to many cases and incidents which the study uncovered.
The assessment of interpretive research differs from positivist theory assessment. Positivists seek rigor using statistical criteria and conceptions of reliability and validity to evaluate the quality of quantitative findings. Sample size, common methods bias and sampling error are common concerns. In contrast, meaning focused research in the interpretive tradition is assessed in terms of trustworthiness criteria including credibility, transferability, dependability and confirmability and authenticity criteria including fairness and ontological, catalytic and tactical authenticity (Guba and Lincoln, 1996: 105). Recent work has shown that the convincingness of ethnographic texts in particular is based on authenticity, plausibility and criticality, where plausibility is defined as the ability of the text to connect the reader and the subject's world, and criticality refers to the ability of the text to probe reader to reconsider taken-for-granted knowledge (Golden-Biddle and Locke, 1993: 599-600). Ultimately, the validity of any interpretive research endeavor is grounded in the return visit, when past research findings are used by yet a next researcher as a guide for entering the field and interacting with the native groups described in the research (Van Maanen, 1981: 27).
The third paradigm of interest, critical postmodernism, is a combination of two somewhat different worldviews -- critical theory and postmodern scholarship. Critical theory is a tradition developed by the Frankfurt School in Germany based on the German tradition of philosophical and political thought stemming from Marx, Kant, Hegal and Max Weber (Kincheloe and McLaren, 1994: 138). Critical theorists departed from Marxist orthodoxy on many issues but maintained a focus on the changing nature of capitalism and the forms of domination, injustice and subjugation capitalism produced.
A basic assumption of the critical tradition is that the material world we encounter is both real and is produced by and through capitalist modes of production. Capitalism contains a basic inequality or contradiction which operates both as a social value and as a social structure: the owner of production (the capitalist) has the right to exploit workers by paying a wage which is less than the economic value which the worker produces. The owner of capital also has the right to appropriate the difference between the value produced and the wage paid. This difference is the surplus labour value of labour and retaining it involves the skimming of profit or surplus labor value by the capitalist or owner of production. This basic contradiction engenders other contradictions and inequalities which make capitalism an inherently exploitive system. These contradictions and forms of exploitation are masked by ideology, a publicly disseminated theory of everyday life events which people use to explain or make sense of events and which encourages people to accept the status quo structures of society as natural, unalterable givens.
Critical scholarship seeks to de-reify the taken for granted as natural structure of capitalism and to uncover forms of inequality and injustice as well as to confront these in ways which transform society into a more democratic institution. Critical scholarship thus seeks to transcend taken for granted beliefs, values and social structures by making these structures and the problems they produce visible, by encouraging self conscious criticism, and by developing emancipatory consciousness in scholars and social members in general (Kincheloe and McLaren, 1994: 140). The goal is social transformation involving the displacement of existing structures of domination, the development of more democratic structures and the opening of opportunities for social participation among persons previously excluded and dominated.
Postmodernism is a form of scholarship which emerged in part through the work of French intellectuals such as Lyotard, Derrida and Foucault. These scholars were well acquainted with critical thought and sensed fundamental limits in critical thought related to changes in the nature and logic of capitalism. Two important changes addressed by postmodernists were the loss of grand narratives and the changing nature of social conditions which make critical reflexivity increasingly difficult to develop and sustain. Critical postmodern research has often focused on discourse at the micro level, in contrast to a somewhat more macro level focus in critical theory research. Critical postmodern research frequently investigates discourse which constructs hidden dichotomies of power and which allows social categories and categorization to operate as hidden mechanisms of control which reproduce status quo structures of domination. The task in critical postmodern analysis has been to deconstruct discourse to reveal hidden structures of domination, particularly dichotomies (e.g. male/female) and then reconstruct or offer alternative, less exploitive social arrangements.
Theories in the critical postmodern tradition take many literary and narrative forms (Boje, Gephart and Thatchenkary, 1996) including historical essays and analysis, field research and case studies. Some critical research also uses conventional positivist methods including survey research. Thus rather than methodological differences it is a commitment to dialectical analysis and to critical/postmodern theory which most clearly differentiates critical postmodern research from positivism and interpretivism. Critical/postmodern research is commonly distinguished by a demonstrated consistency with previous Marxist, critical and postmodern concepts such as alienation, contradictions and commodification. It seeks to provide historical insights including reexamination of important events to reveal previously unacknowledged forms of domination and exploitation. Further, critical postmodern scholarship provides interpretations which allow readers to transcend their narrow, ideologically based views of social phenomena, to de-reify social structures that were previously seen as immutable to change, and to envision and enact social changes to redress basic inequalities and contradiction in capitalist society. Postmodern scholars have given particular attention to the changing nature of signs as fundamental social phenomena which sustain the objective, subjective and intersubjective character of society or social order. Postmodern scholars have investigated how signs saturate our lived experiences yet they are also becoming ever more distant and detached from the things they signify or to which they refer. Deconstruction and textual analysis are two important approaches to investigating and theorizing this breakdown in the relationship of sign to signifier and in addressing the broader social consequences of this fundamental change in knowledge.
Positivism and post-positivism continue to dominate the institutional structure of social and management research. However, the emergence of important limits to positivism and the rise and refinement of well established alternatives to positivism circumscribe real limits to the legitimacy and effective domain of positivist knowledge. Interpretive constructivism offers ways to understand members' own meanings and theories of the world, a fundamental challenge for any scholarly inquiry seeking to have practical relevance. And critical postmodern scholarship challenges the value neutral nature of positivism and even interpretive research. It challenges normal positivist science by displaying that particularistic and elite interests are served by and embedded in positivist knowledge hence positivism serves to reproduce structures of inequality and oppression. Critical postmodernism often borrows methods and theory from interpretive research but it utilizes these in a context where theoretical ideas are used to expose problems of capitalism and to support and encourage political action.
These three traditions are, I argue, the three important and prevailing paradigms or views of the world which are now shaping social, organizational, and management research. At present they are somewhat separate but not greatly distant from one another. Whether these forms diverge, converge for become an integrated scheme for multiple paradigms analysis is perhaps the next important methodological question facing our field.
Berger, P. and Luckmann, T.J. 1966. The Social Construction of Reality. New York: Doubleday and Company, Inc.
Boje, D., Gephart, R. and Thatchenkary, T. 1996. Postmodern Management and Organization Theory. Newbury Park, CA.: Sage.
Blumer, H. 1969. "What is wrong with social theory?" Ch. 8, pp. 140-152 in H. Blumer, Symbolic Interactionism: Perspective and Methods. Englewood Cliffs, N.J.: Prentice-Hall, Inc.
Cicourel, A.V. 1980. "Three models of discourse analysis: the role of social structure." Discourse Processes, 3: 101-132.
Gephart, R.P. 1978. "Status degradation and organizational succession: an ethnomethodological approach." Administrative Science Quarterly, 23 (4): 553-581.
Gephart, R.P. 1993. "The textual approach: risk and blame in disaster sensemaking." Academy of Management Journal, 36(6):1465-1514.
Golden-Biddle, K. and Locke, K. 1993. "Appealing work: an investigation of how ethnographic texts convince." Organization Science, 4: 595-616.
Guba, E.G. and Lincoln, Y.S. 1994. "Competing paradigms in qualitative research". Pp. 105-117 in N.K. Denzin and Y.S. Lincoln (Editors) Handbook of Qualitative Research. Newbury Park, CA.
Jameson, F. 1991. Postmodernism or, the Cultural Logic of Late Capitalism. Durham, N.C.: Duke University Press.
Kincheloe, J.L. and McLaren, P.L. "Rethinking critical theory and qualitative research." Pp. 138-157 in N.K. Denzin and Y.S. Lincoln (Editors) Handbook of Qualitative Research. Newbury Park, CA.
Knorr-Cetina, K. D. 1981. The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science. Oxford: Pergamon.
Miles, M.B. and Huberman, A.M. 1993. Qualitative Data Analysis: a Sourcebook for New Methods. Newbury Park, CA.: Sage.
Schutz, A. 1973. Collected Papers I: The Problem of Social Reality. The Hague: Martinus Nijhoff.
Schwandt, T.A. "Constructivist, interpretivist approaches to human inquiry." Pp. 118-137 in N.K. Denzin and Y.S. Lincoln (Editors) Handbook of Qualitative Research. Newbury Park, CA.
Spradley, J.P. 1979. The Ethnographic Interview. New York: Holt, Rinehart and Winston.
Van Maanen, J. 1981. "Fieldwork on the Beat: An Informal Introduction to Organizational Ethnography". Slide show, cassette tape and transcript prepared for Conference on Innovations in Methodology for Organization Research, American Psychological Association, 1981. First presented to the Center for Creative Leadership, Greensboro, N.C., March 26-27, 1981.
Van Maanen, J. 1988. Tales of the Field. Chicago: University of Chicago Press.
[back to top]