Computational Text Analysis for Public Management Research

Abstract

Organizations produce copious volumes of written documents, including position papers, meeting summaries, minutes from hearings, presentations, and budget justifications. These documents present a wealth of untapped information, which can shed light on a variety of organizational factors-individual and group behaviors, managerial and policy choices, and other key inter- and intra-organizational dynamics that are of great interest to public managers and public administration scholars. Computational text analysis methods offer a highly generalizable means of tapping into these documents in order to generate objective organizational data. We demonstrate a general method for analyzing public texts by applying the Latent Dirichlet Allocation (LDA) approach to measuring budget orientations in county budget documents. LDA is a nonparametric Bayesian method, which is used to extract topical content from collections of documents. We demonstrate how this method can be utilized to measure the functions of budgets in county budget narratives in the state of California, highlighting both within- and between-county variation. This annotated computational analysis of documents is an example of how machine-learning techniques can greatly enhance longitudinal, comparative studies in public management and governance research.