The Social Construction of Organizational Knowledge: A Study of the Uses of Coercive, Mimetic, and Normative Isomorphism

Administrative Science Quarterly, Dec, 1999 by Mark S. Mizruchi, Lisa C. Fein

Arguing that knowledge in the social sciences is socially constructed through the selective interpretation of major works, we examine the fate of a classic article in organizational theory, DiMaggio and Powell's 1983 essay on institutional isomorphism. We show that one aspect of this article, the discussion of mimetic isomorphism, has received attention disproportionate to its role in the essay. A detailed examination of 26 articles in which researchers attempted to operationalize various components of DiMaggio and Powell's model shows that measures used to capture one of their concepts could have served as valid measures of one of the others. Findings show that DiMaggio and Powell's thesis has become socially constructed, as authors have selectively appropriated aspects of the work that accord with prevalent discourse in the field, and that centrally located researchers in sociology and organizational behavior are more likely than other scholars to invoke this dominant interpretation of their article. [*]

Scholars have many motives for their writings. The most obvious and immediate issue is that publishing is the primary indicator of success in one's career. For academic scholars, publication forms the basis of hiring, tenure, and promotion decisions, as well as the basis of one's reputation. Not only must one publish, but one's writings must receive critical acceptance from one's peers. But even more important than critical acceptance are two additional goals: scholars want their work to be read, and they want their work to influence that of other scholars. In the sciences, one's scholarly influence can be gauged by the extent to which other scientists make use of one's theories, hypotheses, and findings (cole, 1992). This influence manifests itself in citations but also in the extent to which one's work is the subject of discussion or even the basis of an article or book itself. At its highest level of influence, one's work forms the basis of a school of thought, a perspective, or approach whose origin is at tributed to the work of a particular scientist (or a small number of scientists).

Whatever the extent of pure curiosity, self-fulfillment, and joy in work that might motivate the efforts of scientists, most hope that their work will be, if not revolutionary, at least influential. But the possibility that one's work will be used by others raises an interesting question: What if that work is misinterpreted or used in a way that differs from the original intent of the author? "Just spell my name right" is a common plea among those striving for careers in music and theater, indicating that one's primary concern is to be known, regardless of the basis of that knowledge. The sciences are not the music industry, however, and it seems reasonable to assume that a scholar's work, when used by others, will be accurately interpreted. But we know that this is not always true. Latour (1987: 40) noted that "a paper may be cited by others ... in a manner far from its own interests" and even "to support a claim which is exactly the opposite of what its author intended." At the same time, classic works are frequently described as often cited but rarely read. This accounts for the surprise that readers often experience when they actually go back and read such works.

The fact that classic works in a field are often cited and discussed without being carefully read (or read at all) suggests the possibility that these works can become social constructions, taking on identities created as much by their users as their authors. However frequent or infrequent such events may be in the natural sciences, they are far from rare in the social sciences. [1] Examples of such socially constructed reinterpretations of classic works abound. As Adatto and Cole (1981: 151) noted in their discussion of citations to classic works, such references "selectively present knowledge that fits in with the currently held paradigms." Virtually any work in which an author reexamines classic texts will reveal a series of widely assumed truths that turn out to be false. These "truths," similar to what Merton (1987) called pseudofacts, often form the basis of subsequent works, leading to even further distortion of the original classic (Cole, 1975: 212-213; Small, 1978: 338; Cronbach, 1992; Hamilton, 1996 ).

In this paper we argue that the interpretation and uses of knowledge have a socially constructed character and that this can lead organizational researchers, as well as scholars in general, toward misleading representations of phenomena. We illustrate this argument by examining the uses of a highly influential article in organizational theory: DiMaggio and Powell's (1983) piece on institutional isomorphism. We show that although there are several components to DiMaggio and Powell's argument, one aspect of their discussion has received disproportionate attention--at the expense of other, equally prominent formulations--which we believe can be accounted for by the extent to which this component corresponds with prevailing discourse in organizational theory.