Estimating Party Positions across Countries and Time—A Dynamic Latent Variable Model for Manifesto Data

Abstract

This article presents a new method for estimating positions of political parties across country- and time-specific contexts by introducing a latent variable model for manifesto data. We estimate latent positions and exploit bridge observations to make the scales comparable. We also incorporate expert survey data as prior information in the estimation process to avoid ex post facto interpretation of the latent space. To illustrate the empirical contribution of our method, we estimate the left-right positions of 388 parties competing in 238 elections across twenty-five countries and over sixty years. Compared to the puzzling volatility of existing estimates, we find that parties more modestly change their left-right positions over time. We also show that estimates without country- and time-specific bias parameters risk serious, systematic bias in about two-thirds of our data. This suggests that researchers should carefully consider the comparability of party positions across countries and/or time.

Partisanship, Ideology, and Representation in Latin America

“…In this respect, this paper is closely related to Jessee (2010), Shor and Rogowski (2010), Tausanovitch and Warshaw (2013), and Battista, Peress and Richman (2013), who use common items and scaling techniques to examine policy representation in the United States. In addition, my study nicely aligns with recent analyses of Europe's common ideological space by Bakker et al (2011), König, Marbach and Osnabrügge (2013), and Lo, Proksch and Gschwend (2014), as well as Saiegh (2009) who uses scaling techniques to estimate the location of political actors in Latin America.…”

“…Relying on European political groups to identify the location of parties in different European countries is another case in point (König, Marbach and Osnabrügge, 2013;Lo, Proksch and Gschwend, 2014). An alternative approach is to treat questions that are asked in the same form to respondents in different countries as bridges to create a common spatial map.…”

Towards the median economic crisis voter? The new leftist economic agenda of the Front National in France

“…This cluster concerns only a minority of the FN's policies and is therefore not included in this analysis. political text coded into discrete categories (König, 2013). Here we employ the additive index based upon the economic policy scale constructed by Benoit and Laver (2007, p.100) 5 .…”

Evaluating and Improving Item Response Theory Models for Cross-National Expert Surveys

“…1 Multiple raters may have quite di↵erent understandings of latent concepts and this variation in conceptualization may be especially exacerbated when experts hail from varying cultural and educational traditions. Indeed, in the context of party manifestos, Konig, Marbach & Osnabrugge (2013) show that expert ratings exhibit substantial cross-national bias. More generally, people with varying backgrounds are likely to apply di↵erent standards when rating concepts (King & Wand 2006).…”

“…Clinton & Lapinski 2006, Clinton & Lewis 2007, Treier & Jackman 2008, Pemstein, Meserve & Melton 2010, Linzer & Staton 2012, Konig, Marbach & Osnabrugge 2013, Schnakenberg & Fariss 2014, Fariss 2014). These models relax two key assumptions implicitly made by traditional approaches to aggregating ratings in expert surveys.…”