Philip Warncke

Philip Warncke

Post-doctoral fellow in Political Science

Free University of Berlin

Welcome to my Website!

I am a comparative political psychologist studying the origins, properties, and consequences of political belief systems. I also develop conceptual and computational methods for latent constructs – like belief systems – that are often tricky to capture with conventional statistics. My teaching interests are broad and I particularly enjoy interactive, discussion-based seminars that challenge students to think outside the box. I currently work as a Pre-/Post-doctoral fellow at the SCRIPTS Data and Methodology Center at Freie Universität Berlin and have previous work experience as a statistics and social data science consultant at the Odum Institute. You can find my research, CV, links to software & tutorials, and teaching material on these pages. Please feel free to contact me with any inquiries, bug reports, or based on substantive research interests.

Interests

  • Ideologies and mass belief systems
  • Network-based, latent space models
  • Comparative public opinion
  • Data visualization

Education

  • PhD in Political Science, 2024 (May)

    University of North Carolina at Chapel Hill

  • MA in Political Science, 2020

    University of North Carolina at Chapel Hill

  • MSc in Comparative Politics, 2015

    London School of Economics and Political Science

  • BA in Political Management, 2014

    Hochschule Bremen

Publications

Active, assertive, anointed, absconded? Testing claims about career politicians in the United Kingdom

We undertake a comprehensive investigation into several common critiques of career politicians. Career politicians are said to be self-serving. Active and assertive when it suits their career interests, and much more interested in attaining higher offices than in serving as constituency-oriented MPs. Yet, empirical investigations of their alleged behaviours are few, and the results are patchy and mixed. Focusing on the UK case and using a multi-dimensional conceptualisation that accords with academic and popular understandings of career politicians, our paper draws on uniquely rich attitudinal and longitudinal behavioural data covering the first large generational wave of career politicians to be elected to parliament in the early 1970s. It reports findings consistent with contemporary critiques, suggesting that such dispositions are inherent in the role of career politician. The strongest career politicians among this first wave concentrated strategically on career-serving activities, voted strategically to safeguard their careers, attained and retained successfully ministerial offices, and prioritized their personal goals over their party obligations. The paper further demonstrates that different measures used by researchers can produce contradictory results, and that future comparative research should seek to range beyond unidimensional indicators.

Work in progress

Boundless but Bundled: Modelling Quasi-infinite Dimensions in Ideological Space

Ideological scales derived from policy position items are prevalent in political psychology and behavioral research. However, the underlying spatial assumptions of these scales are rarely scrutinized. This study investigates how assumptions about the dimensionality of the latent ideological space can significantly impact empirical estimates, even when using the same data from the same respondents. Through a comprehensive literature review and statistical simulations using data from the ANES, we demonstrate that the optimal number of latent ideological dimensions increases without bound as researchers include additional items for analysis. At the same time, nearly all latent ideological factors found within same attitude set are sizably and positively correlated with one another. In light of these findings, we propose an alternative modeling framework that seeks to reconcile unidimensional and multi-dimensional aspects of mass ideology. Our Bayesian hierarchical latent variable approach simultaneously estimates mass ideology as a higher-level, uni-dimensional expression of correlated, lower-level, multi-dimensional building blocks. This approach enables researchers to assess whether particular socio-demographic or psychological predictors, such as income, gender, or egalitarianism, are consistently related to specific sub-dimensions (e.g., economic, socio-cultural, racial ideology) or instead a generalized, uni-dimensional representation thereof. Our results underscore the potential value of this approach, offering insights into the unique characteristics of different ideological factors and their overarching parent dimension.

The Hidden Problem in Big Data: Even Infinite Information does not Guarantee Consistent Measurement

The social sciences heavily depend on the measurement of abstract constructs for quantifying effects, identifying association between variables, and testing hypotheses. In data science, constructs are also often used for forecasting, and, thanks to the recent big data revolution, they promise to enhance their accuracy by leveraging the constantly increasing stream of digital information around us. However, the possibility of optimizing various social indicators implicitly hinges on our ability to reliably reduce complex and abstract constructs (such as life satisfaction or social trust) into numeric measures. While many scientists are aware of the issue of measurement error, there is widespread, implicit hope that access to more data will eventually render this issue irrelevant. This paper delves into the nature of measurement error under quasi-ideal conditions. We show mathematically and by employing simulations that single measurements fail to converge even when we have access to progressively more information. Then, by using real-world data from the Social Capital Benchmark Surveys, we demonstrate how the addition of new information increases the dimensionality to the measured construct quasi-indefinately which further contributes to measurement divergence. We conclude discussing implications and future research directions to possibly solve this problem.

Using Particle Physics to Estimate Non-linear Expressions of Latent Constructs

Political scientists often use survey instruments to indirectly infer latent constructs such as social trust, political efficacy, and ideology. However, traditional techniques used to reconstruct latent quantities from survey data, such as principal components, factor analysis, and item response theory models, have limitations as they assume linearity among the underlying co-dependencies among a given set of survey items. This assumption does not hold for complex nonlinear expressions of latent constructs, such as when survey respondents from opposite ends of the ideological spectrum share a stance on a particular issue but for completely different reasons. For example, a left-leaning respondent may see cash benefits for child care as an important instrument for gender equality while conservatives welcome incentives for women to stay home and give birth to more children. To liberals, GMO’s might appear as environmental hazards while conservatives reject them as they could tamper with God’s eternal creation. To address such non-linearities, we propose a new method that models survey item-responses as force-directed statistical networks. By treating item responses as physical particles in finite space, a force-directed algorithm can detect complex, non-linear relationships within latent constructs. Generalized additive models and LOESS-fit models applied to the spatial equilibrium of item response particles can subsequently be used to calculate sparse, (non-)linear latent factors that optimally characterize this space. This method is flexible, allows for multiple scoring options, and outperforms traditional latent variable models in the presence of non-linear expressions of the underlying latent construct.

Software links & tutorials

ResIN for R: Response Item Networks package for R

Method. CRAN. GitHub. R-Package tutorial.

SmashingPumpkins for R: Color Palettes for Data Visualization

GitHub.

Teaching resources

ResIN Workshop: Response Item Networks

Bigsss Computational Summer School 2023. Slides. Solution.

POLI 215: Political Psychology: An Introduction

UNC Chapel Hill, Spring 2022. Syllabus.

POLI 236: The Politics of East-Central Europe

UNC Chapel Hill, Fall 2022. Syllabus.

POLI 209: Analyzing Public Opinion

UNC Chapel Hill, Spring 2021. Syllabus.

Contact