In my studies as a doctoral student of political science at Cologne University, I focus on Populism. Specifically, my doctoral dissertation analyses cross-national differences with respect to populist proclivities, as well as scrutinize the validity and psychometric properties of Populist Attitudes instruments. My doctoral dissertation is divided into three main projects (slides):
Populism is on the rise in Europe. In the past decade – with the emergence of more volatile electorates and protracted global economic crisis – radical and right-wing populist parties have garnered unprecedented political power at national and supra-national elections. In order to better understand this phenomenon, the we focus on the measurement and empirical evaluation of populist attitudes, which are thought to explain – at least, partially – predilections for Populist candidates and parties. Indeed, as of recent, a number of prominent studies have converged on a set of six items which have been widely used to gauge this latent construct. However, despite a vast literature, the psychometric properties of this instrument remain largely assumed, rather than empirically tested. For our part, and as a mean to further validate the populist attitudes scale, two original items were developed and added to the item battery of the LIVEWHAT project. In 2015, the resulting scale surveyed populist attitudes across nine European countries (n = 18370). The objective of this project is three-fold. Drawing insights from both classical test and item response theories, the first objective is to provide an in-depth assessment of the scale’s psychometric properties, such as internal consistency and dimensionality. Secondly, evaluate the instrument’s (and its items) ability to differentiate among respondents across the latent construct. And thirdly, analyse and compare cross-national idiosyncrasies. Results indicate the instrument is structurally robust, however, while the scale was conceived to measure populist attitudes across its entire continuum, the populist instrument is informative only at a limited range of the theorized latent construct. Implications to the construct- and predictive-validity are discussed, and recommendations for future operationalisations are suggested. In terms of cross-national comparisons, the nine surveyed countries can be categorized into two sub-groups: Italy, Spain, Greece, France and Poland, which showed high levels of populism endorsement; and Sweden, Switzerland, Germany and the UK, which consistently exhibited lower populist proclivities.
For more information you can read the first article draft, prepared for the 2016 MPSA, here.
In Europe and in the Americas, populism has eroded traditional political cleavages. This, with the creation of new issue dimensions, affected people's political preferences which ultimately brought about a Populist zeitgeist. However, the majority of populism research pays particular attention to the supply-side of politics - i.e., its definition and associations at the macro-level. Here, we take a different approach: to explore and concentrate on the demand-side of populism: e.g., micro-level associations and analysis via latent variable modeling. It is true, however, that scholars are gradually shifting towards a more balanced approach, with scholars consistently converging around a Populist Attitudes battery originally designed by Hawkins and Riding (2010). And even if there are several variations of this battery being used to measure populist attitudes in both sides of the Atlantic, there is still no systemic and in-depth latent construct analysis for these survey items. This is the goal of this project, to validate and generalize our previous findings through the inclusion of several data sources across the Americas (AmericasBarometer, CCES, UCEP) and Europe (LIVEWHAT, ESS). Altogether, the scope of analysis is significantly broadened, and an evaluation of populist attitudes at the continental, cross-regional and cross-country level will be provided. Additionally, a detailed psychometric evaluation is carried out in terms of the Populist instrument's (and its items') performance. An initial assessment confirmed previous findings in that the Populist Attitudes scale fails to capture the full range of the latent populist construct, remaining limited to its ‘moderate’ spectrum. Therefore, we recommend extensive scale development (designing, testing, and validation) of new items as to increase the accuracy with which the instrument is able to capture populist attitudes.
This project will figure on the new book of Kirk Hawkins on Populism.
Political science researchers - particularly those in comparative perspectives - often rely on individual-level data as a mean to put theories involving non-observable constructs to a test (e.g., attitudes, trust, political orientation). Prominent examples such large-scale cross-national survey data are: Eurobarometer, American National Election Studies, European Social Survey, Comparative Study of Electoral Systems, International Social Survey Programme, and the World Values Survey. However, in using comparative sources to model and infer with latent constructs, one aspect is seldom addressed: the issue of assumed comparability. The idea here is that comparative research results assume implicitly that individuals living in widely varying contexts have an equal understanding of what is asked by a given question. By ignoring to consider measurement non-equivalency - i.e., systematic distortion in response behavior of individuals from different countries - comparative research risk to attribute differences in response behavior, which are in fact not due to true attitudinal differences but the result of country specific (i.e., non-random) measurement error, to true effects. In other words, when this assumption is violated, the repercussions are consequential to both interpretation and validity of inferential comparability. The current project attempts to correct this shortcoming. It is intended to contribute to the discussion of latent variables in political science by (a) reviewing the literature on measurement equivalence; (b) to formalize an intuitive and general guideline for checking for measurement equivalence; (c) to do so comparing two techniques - differential item functioning and multi-group confirmatory factor analysis - and discuss its pros and cons of each approach.
Projects I participate include
Other "working in progress" scientific projects