The term “populism“ is pervasive in today’s political debate, but there is little agreement on how to actually define or measure populism. For example, are Vladimir Putin or Recep Erdogan populists or not? Do Emmanuel Macron or Theresa May rely on populist strategies? And how can we capture populist attitudes of voters, in order estimate the voting potential of populist parties?
The goal of this project is to kick-start a cooperation between GESIS and the Kiel Institute and to submit a joint, interdisciplinary, and methodologically innovative grant application. The grant will aim to produce new instruments to measure populism - both on the micro-level (in polls) as well as on macro-level (governments, politicians). We are convinced that a clean quantification of populism will go a long way for understanding the drivers and consequences of this phenomenon.
Part 1 (GESIS) draws on methods and theories from political science and adopts a micro perspective. Who votes for populist parties and in which context? Is there a “populist mindset”? An important limitation in the literature today is that there are no standardized methods to capture populism in political surveys. We therefore plan to compare the populism scales and measurement approaches of established researchers (e.g., Hawkins, Riding, Mudde, Akkerman, Zaslove) with the approach of the GESIS Panel (2016), ALLBUS (1988) and the European Value Study. On this basis, our goal is to develop new, standardized, and internationally comparable survey instruments, that will be made available to the entire research community.
Part 2 (IfW Kiel) adopts a macro perspective and will be implemented by a team of economists. We aim to create a new “consensus-based” and continuous populism index for political leaders and parties worldwide. The idea is straightforward: If the bulk of scientists and commentators call a politician a populist, it is almost certain that this politician is indeed a populist. By contrast, if this description is rarely used in connection with a politician, we assume that she/he has only slight or no populist tendencies. To quantify the number and intensity of populist attributes we rely on modern tools for text analysis. Specifically, the “Consensus-Based Index of Populism” will be measured in two ways: 1) Via a systematic, quantitative evaluation of the entire scientific literature on populism (about 150 books and 500 articles). 2) By evaluating hundreds of thousands of print media articles (e.g. in Factiva) with machine learning tools (for example “how often are the name Berlusconi and the term populism mentioned in the same sentence?”). The results of both approaches will then be compared in order to assign a populism index value (ranging from 0 to 100) for every political leader or party.
05/2017 - 07/2017