Pugh 1969 - An Empirical Taxonomy of Structures of Work Organizations

Google Scholar Link

Pugh, D. S., Hickson, D. J., & Hinings, C. R. (1969). An empirical taxonomy of structures of work organizations. Administrative science quarterly, 115-126.

Todo: insert figure 3

This paper is an interesting example of “you thought it was one thing, but actually it’s many things.”

Summary

Information from 52 organizations, regarding five structural variables, was subjected to factor analysis, revealing three dimensions: structured vs unstructured activities, concentrated vs dispersed authority, and line control vs impersonal control of workflow. These three dimensions were used to elaborate a taxonomy of bureaucracy, which were also revealed visually as clusters. Since most of my factor analysis experience revolves around psychological scales, this was a novel use case for me. It might be fruitful to treat seemingly unitary constructs to factor analysis to determine if there are patterns of covariance, which would suggest multiple dimensions. I imagine that more modern clustering recognition algorithms could be used for more precise results, especially in analysis of more than three dimensions.

todo: insert fig. 3

Highly structured organizations with low concentrations of authority are called workflow bureaucracies. Examples include large manufacturers. Organizations with high concentration of authority and low structure of activities are called personnel bureaucracies. In these, the daily work is not bureaucratized, but everything related to employment is. Examples include central government departments. Organizations that are low on both structure and authority concentration are called implicity structured organizations. These tend to be small organizations, run by an owner operator. Though quite rare, a full bureaucracy is one in which both structure and authority concentration are high. Including the third dimension of line control vs. impersonal controls implies a progression amongst organizations. As organizations grow in size, they increasingly become structured (moving from a nascent full bureaucracy to a full bureacracy). As organizations increasingly rely on mechanization to produce standardized products, control shifts from workflow personnel to impersonal means, such as processes and new specialists. Movement along this axis progresses from a preworkflow bureaucracy to a nascent workflow bureacracu and finally to a workflow bureaucracy.

Application

Factor analysis has applications outside of the strictly psychological.