Machine Learning for Public Administration Research, with Application to Organizational Reputation

Machine learning methods have gained a great deal of popularity in recent years amongpublic administration scholars and practitioners. These techniques open the door tothe analysis of text, image and other types of data that allow us to test foundationaltheories of public administration and to develop new theories. Despite the excitementsurrounding machine learning methods, clarity regarding their proper use and potentialpitfalls is lacking. This paper attempts to fill this gap in the literature throughproviding a machine learning “guide to practice” for public administration scholars andpractitioners. Here, we take a foundational view of machine learning and describe howthese methods can enrich public administration research and practice through theirability develop new measures, tap into new sources of data and conduct statisticalinference and causal inference in a principled manner. We then turn our attentionto the pitfalls of using these methods such as unvalidated measures and lack of interpretability. Finally, we demonstrate how machine learning techniques can help us learn about organizational reputation in federal agencies through an illustrated example using tweets from 13 executive federal agencies. All R code, analyses and data described in this paper can be found in the Online Appendix