What the Annual Reports of a National Ombudsman Institution Tell Us

On the Use of Text Mining in the Empirical Sociology of Law

Abstract

This article assesses the possibility of using text mining in socio-legal research by analyzing the discursive self-presentation of the Austrian national ombuds institution in a corpus of annual reports. Using the program Alceste, the paper examines how topics are prioritized in the description of activities and how the institutional mission is presented and legitimized. Alceste identifies a total of nine lexical worlds, which point to the diversity of roles exercised by the ombudsmen and define three ideal types of approaches to the ombudsman's office. The text highlights that these different, partly interrelated role enactments are characteristic of certain individual organizational units and persons, which indicates a personalization of the office, and that some of them are shared by several ombudsmen.