Worker Interdependence and Output: The Hawthorne Studies Reevaluated

Recent work has called into question the "human relations" interpretation of the evidence from the classic Hawthorne studies, arguing that external factors suffice to explain the variation in workers' output found at the Hawthorne plant. In contrast, this paper demonstrates that, allowing for a wide range of other factors mediated by the social interactions in the working group, workers' productivity levels in the Hawthorne plant were indeed interdependent. Using five years of weekly data from the original Hawthorne studies, a structural model of worker productivity is estimated. This model allows for interdependence of workers' output levels, while recognizing that, although mediated by small group interactions, external and experimental variables can also play a role. The results reveal the joint importance of external factors and worker interaction in affecting the level and variability of output. The human relations approach to industrial sociology is not controverted by the original Hawthorne data from which it began.