Isomorphism in Organizational Language

This page gives an overview of my research on the phenomenon I call "organizational language," a ubiquitous part of organizational life. Organizational language is the discourse organizations produce to communicate with their internal and external audiences. My thesis is a detailed examination of the organizational language in texts and documents intended for external audiences. Organizations constantly produce and consume these texts and documents, and they are a primary way that organizations interact with other organizations and institutions, but they are often taken for granted in organizational research. I argue we can gain fundamental insights into how institutions influence organizations and how organizations respond to and are shaped by institutional forces by closely studying organizational language. With a detailed analysis of how organizations interact with institutions through products such as texts and documents we can improve our understanding of how organizations translate institutional forces into a local context and our measures of institutionalization.

One manifestation of the institutionalization process that is of particular interest to organizational researchers and that they have theorized should be present in organizational language is isomorphism, or similarity. Typically, isomorphism has been treated as a black box. I analyzed nearly 300 texts in 10 different datasets and from two different types of organizations (universities and corporations), and clearly establish how organizations translate pressures from their environment into organizational language. Variation in the extent of isomorphism is predictable based on 1) how susceptible a type of organization is to its environment, 2) the nature of who the text speaks for (an individual or the organization), and 3) the primary motivation of the organization for producing the text (asserting a distinct identity or displaying responsiveness to their audiences). More susceptible organizations (universities) display a lower extent of isomorphism in their texts and greater responsiveness to their audiences. Texts that speak for the organization display greater isomorphism, but focus more on asserting a distinct identity. In addition to developing novel measures of isomorphism in texts, I break up isomorphism into a more multifaceted concept, incorporating organizational context and authorial motivations, and pointing to the key role that a text's implicit audiences have in encouraging isomorphism.

The measures of isomorphism I develop include studying the networks of concepts in the texts using automated text analysis tools. The two pictures below demonstrate the different approaches of universities and corporations to the same type of text: privacy policies. The privacy policies from universities display less consensus about which concepts to include in their texts, leading to a sparser network compared to the corporate policies. While both types of policies use concepts related to responding to their environment (in red), university policies use many more of these concepts compared the the corporate policies, which use a mixture. Rhetorically, this is because the university texts focus on the concept of privacy as the crux of the text: they explain it, define it, and constrain it. In contast, the corporate policies focus on the relationship between the organization and the reader (us/we vs. you), but balance an interest in privacy with other concepts such as serving the customer. Reflecting the overall findings, these two pictures illustrate how those organizations that are more susceptible to their environment (universities) do display responsiveness, but with less isomorphism across texts.

University Privacy Policies	Corporate Privacy Policies