RESEARCH ISSUESSHAWNEE VICKERY, Feature Editor, Eli Broad Graduate School of Management, Michigan State University DIVERSITY IN RESEARCH ON INFORMATION SYSTEMS by Daniel Robey
Research on computer-based information systems currently reflects more diversity of research practices and methods than any other field in business administration. Research in information systems draws upon organizational science, computer science and engineering, management science, and many other disciplines in pursuit of knowledge about the design, development, implementation, use, and consequences of information technology applied to business practice. For some, such diversity signals confusion and a need to return to more fundamental scientific objectives. These critics claim that the collective body of research in information systems lacks coherence and that journals in information systems are schizophrenic compendia of unrelated research reports. For others, diversity signals health, vitality, and a chance that research in information systems may actually contribute to the solution of applied business problems. This article addresses the issue of diversity in the conduct of research in the field of information systems. Managing and valuing diversity in the workforce has quickly become one of human resource management's primary interests in the 1990s. Valuing diversity is not only "politically correct" but also good business, given the changing demographics of the workforce and the marketplace. Is it possible that diversity also holds the key to more productive research in information systems and the decision sciences? Not long ago, researchers studying information systems seemed eager to establish a new scientific paradigm to help define their fledgling discipline alongside the more established disciplines in business schools. Lately, this search seems to have ended with the resolution by researchers to draw guidance from relevant "reference disciplines." In a recent issue of Information Systems Research, Senior Editor John King explained that the diversity in research in information systems is traceable to our origins as "an intellectual convocation that arose from the confluence of interests among individuals from many fields who continue to pledge allegiance to those fields through useful ties of various kinds" (King 1993: 293). Such diversity, King argued, is a source of strength and vitality for our field, not a sign of weakness. In the near future, information systems research will probably reflect even greater diversity by drawing from disciplines such as political science, law, and architecture. Like it or not, diversity will continue to characterize research efforts in information systems. The openness of research communities to diversity is evident in fields other than information systems. For example, Organization Science was launched in 1990 with a lead article by the editors entitled, "Can Organization Studies Begin to Break Out of the Normal Science Straitjacket?" (Daft & Lewin 1990). The physical sciences have actually been wrestling with such issues for most of the twentieth century. Quantum mechanics, which challenged Newtonian positivism, was originally advanced in 1900, and Einstein's theory of relativity dates to 1905. The social and applied sciences (including business), which deal with ostensibly more slippery phenomena, should be even more keen to embrace diversity in research. EPISTEMOLOGICAL ASSUMPTIONS One fundamental source of diversity is the differing epistemological assumptions made by researchers. Both interpretive and positivist research can be found in information systems. Interpretive research is work that adopts the epistemological assumption that reality is subject to multiple interpretations. Stated more assertively, reality consists of subjective interpretations and cannot be studied objectively. The purpose of interpretive research is to understand and analyze subjective interpretations and their consequences. This understanding and analysis are, of course, mediated by the researcher, who is advised to acknowledge his or her own subjective frame of reference in reporting findings. By contrast, positivist research adopts the assumption that reality is objective and observable. Researchers are advised to follow the scientific method, which has as its objective the elimination of subjectivity from the research process. Positivist researchers are assumed to be detached, external observers and reporters of what they see. These distinctions are elementary ones that should be part of every researcher's basic training. What is troubling is the argument that either interpretivism or positivism is superior to the other, or that one is valid for research while the other is not. Positivism is often characterized as a dominant paradigm in Western scientific thought, but dominance does not imply correctness. Scientific paradigms reflect the shared assumptions and practices of a community of scientists in a given place and time. Scientific revolutions and paradigm shifts occur, disturbing temporary notions of truth and the conduct of research. In my view, knowledge is generated through academic inquiry that adopts either epistemological assumption. Taken seriously, interpretivism is not a subversive threat to positivism but rather a useful alternative set of guiding assumptions for empirical research. In my own research, I have attempted to understand the wide range of social consequences that ensue when information technologies are introduced into organizations. Positivist assumptions have directed my attention to the objective properties of the technology and guided my search for predictable impacts of technology. Interpretivist assumptions have directed me toward the subjective social meanings that information technology acquires as it is developed and used. As designers, managers, and users interact, shared and competing interpretations of technology emerge which ultimately produce important social consequences, such as the use or nonuse of systems. I see no inherent contradiction in the same researcher performing both interpretive and positivist research. Selection of epistemological assumptions should depend on their usefulness to the particular research questions being addressed. By sharing my assumptions and methods with the community to whom the research is directed, I allow all readers to form their own opinions about the appropriateness of the research for their own purposes. THE RULES OF FREEDOM Research communities long ago established the principle of academic freedom because they realized that the advancement of knowledge thrives on a diversity of ideas and open communication. Academic freedom guarantees scholarly diversity and rejects intellectual hegemony. However, academic freedom is governed by rules that ensure some conformity to fundamental standards of research. Two of these rules are worth emphasizing here. FULL DISCLOSURE OF RESEARCH METHODS. The purpose of the Method section in standard research papers is to disclose the methods used by the researcher. A reader should gain enough information from the description of method to be able to replicate the study with another sample or site. Failure to state the details of method leads critics to dismiss such work as sloppy and unfounded. Research also requires investigators to be honest about the entire research process, disclosing how all data were generated and analyzed and stating the foundation for all interpretations of the data. Research is a process of communicating with a community of scholars, not an isolated activity. If our research papers are unclear or inconsistent on matters of method or interpretation, the members of this community have the right to insist on clarification. Without full disclosure in research, it is impossible to produce contributions to collective knowledge that are valued by the wider community. CONDUCT. Academic freedom means that diverse views are not only tolerated but encouraged and respected. It is common, however, for political motives to compromise academic freedom. For example, less powerful individuals like doctoral students and untenured faculty may be intimidated by senior professors with personal agendas. While politics can never be eliminated from organizations, the loss of academic freedom is too high a price to pay to satisfy self interest. Accordingly, rules of ethical professional conduct must be understood and shared so that the merits of different theories, analyses, and assumptions can be discussed within a climate of trust and respect. We have a right to debate with our colleagues and students and to challenge their ideas. But we also have the responsibility to accept the explanations given and to avoid intimidating our colleagues and students. In conclusion, the diversity that is so apparent in research on information systems is a sign of academic vitality and freedom, not evidence of confusion and conflict. I hope the points raised here remind us about the importance of maintaining a viable forest while we bicker about whose trees are taller, stronger, and prettier. REFERENCES
Daft, R.L., and Lewin, A.Y., "Can Organization Studies Begin to
Break Out of the Normal Science Straitjacket? An Editorial Essay,"
King, J.L., "Editorial Notes,"
DANIEL ROBEY is Professor of Information Systems at Florida
International University. He earned his doctorate in 1973 from Kent
State University and has served on the faculties of The University
of Pittsburgh, Marquette University, Gannon University, and the
Copenhagen School of Economics and Business Administration. He is
the author of three books and numerous articles in such journals as
|