ABSTRACT
This chapter presents the results of textual analysis applied to documents draft by party factions, in order to estimate their policy positions. The data collection process is explained, paying attention to the importance of party congresses as a source of data on intra-party politics. The chapter briefly describes Wordfish, the technique of automated text analysis used in the book, and then presents the estimates related to 254 factions that competed in 83 congresses of 19 Italian parties (1946–2010), along with the estimates of 151 party factions, concerning 39 congresses of 5 French parties (1971–2016), and those of 33 subgroups belonging to the two main German parties (1985–2018). It concludes by proving the validity and reliability of these estimates, as well as the robustness of the results of textual analysis to alternative pre-processing strategies.
