Working Papers

Measurement Uncertainty in Spatial Econometric Models: A Bayesian Dynamic Measurement Model

(under review)

According to theoretical models of political competition, parties strategically adjust their ideological positions to movements made by rival parties. Spatial econometric models have been proposed to empirically model such interdependencies and to closely convert theoretical expectations into statistical models. Yet, these models often ignore that the parties' ideological positions are latent variables and, as such, accompanied by a quantifiable amount of uncertainty. As a result, the implausible assumption of perfectly measured covariates impedes a proper evaluation of theoretical propositions. In order to bridge this gap between theory and empirics, the present work combines a spatial econometric model and a Bayesian dynamic measurement model. The proposed model accurately accounts for measurement uncertainty and simultaneously estimates the parties' ideological positions and their spatial interdependencies. To verify the model’s utility, I apply it to recorded votes from the sixteen German state legislatures in the period from 1988 to 2016. While exhibiting a notable degree of ideological mobility, the results indicate only moderate spatial dependencies among parties of the same party family. More importantly, the analysis illustrates how measurement uncertainty can lead to substantively different results which stresses the importance of appropriately incorporating theoretical expectations into statistical models.

Domestic Issue Competition and the Transnational Diffusion of Electoral Strategies

(under review)

How do parties decide whether to campaign on consensual or conflictual issues, given high degrees of uncertainty? Assuming that parties strategically attempt to attract voters by writing manifestos, I argue that parties use the electoral performance of other parties as heuristics to mitigate uncertainty and to systematically learn about the most promising electoral strategy. The empirical analysis employs tools from spatial econometrics and analyzes party manifestos with regards to the parties' emphasis on the environment – a valence issue with a niche party being the unequivocal issue owner – in 25 EU countries from 1975 to 2015. The results reveal non-trivial interdependencies among European party systems stemming from direct transnational dependencies and indirect spillover effects among electoral strategies. The analysis further identifies conscious learning rather than mere imitation or independent decision-making as the mechanism underlying the diffusion of electoral strategies across national borders. Yet, in line with saliency-based theories, electoral competition mutes the diffusion of strategies at the domestic level. Here, only the success of issue owning green parties exerts a non-linear effect on the other parties' likelihood to address environmental issues. The results have important implications for understanding the dynamics of party competition, learning and diffusion processes, and parties' strategic behavior.

Wheeling and Dealing Behind Closed Doors: Estimating the Causal Effect of Transparency on Policy Evaluations Using a Survey Experiment

(with David Hilpert; under review)

The Transatlantic Trade and Investment Partnership (TTIP) between the EU and the United States is highly technical. Still, the negotiations triggered large-scale protests among citizens with very diverse socioeconomic backgrounds. Why has a complex issue such an enormous mobilizing effect although the economic consequences are either unclear or favorable for the participating regions? We argue that the amount of information available during the negotiations – their transparency – not only shapes the legitimacy of the decision-making process but also affects public evaluations of the final policy outcome. Conducting a survey experiment, we show that non-transparent decision-making decreases citizens' appraisal of the outcome independent of its content. Our findings have important implications for questions of institutional legitimacy and public policy evaluations.

Die „Sonntagsfrage“, soziale Erwünschtheit und die AfD: Wie alternative Messmethoden der Politikwissenschaft weiterhelfen können

(with Thomas Gschwend and Roni Lehrer; under review)

Die Wahlabsichtsfrage, populärwissenschaftlich auch als „Sonntagsfrage“ bezeichnet, wird kritisiert, weil mit ihr der Stimmenanteil der Alternative für Deutschland (AfD) nicht valide zu messen ist. Wir argumentieren, dass alternative Messinstrumente, die Verzerrungen aufgrund von sozialer Erwünschtheit berücksichtigen, besser geeignet sind. Dazu testen wir erstmalig drei alternative Messmethoden - das doppelte Listenexperiment, die Randomisierte-Antwort-Technik und die Weisheit-der-Vielen-Methode - hinsichtlich des geschätzten AfD-Stimmenanteils und vergleichen sie mit der klassischen „Sonntagsfrage“. Unsere Ergebnisse zeigen, dass insbesondere die Weisheit-der-Vielen-Methode eine kostengünstige und gute Erweiterung der politikwissenschaftlichen Fragebatterie ist.

The Wisdom of Crowds Design for Sensitive Survey Questions

(with Roni Lehrer and Thomas Gschwend; under review)

Survey research on sensitive questions is challenging because respondents often answer untruthfully or completely refuse to answer. Existing survey methods avoid social desirability bias by decreasing estimates' efficiency. We suggest the Wisdom of Crowds survey design that avoids bias without decreasing efficiency. We compare the design's estimate of a right-wing populist party's vote share to the real election results in the 2017 German federal election. We find that the Wisdom of Crowds design performs best in terms of both bias and efficiency while other techniques for sensitive questions perform less well. We conclude that the Wisdom of Crowds design is an important addition to the social scientists' survey methodology toolbox.

W(ho Knows)? A Bayesian Approach to Network Uncertainty

Spatial econometric models become increasingly popular in various subfields of political science. However, a prevalent criticism of this class of models is that the validity of the models' inferences are conditional on the pre-defined connectivity matrix, denoted W, which specifies the underlying network of dependencies. At the same time, the true dependence structure is rarely known a priori and theories provide only insufficient guidance which complicates the specification of W. The present study advocates Bayesian model averaging (BMA) as a superior approach to this problem. It also provides practical guidance for political scientists using spatial econometric techniques on how to address network uncertainty. In contrast to mere selection procedures, BMA directly accounts for network uncertainty, which is a special case of model uncertainty that arises from uncertainty about the specification of W, in a coherent framework. Three replication studies demonstrate the benefits of this approach for practical research. The results show that, while model uncertainty is a valid concern, network uncertainty does not dramatically change the substantive conclusions derived from spatial econometric models, especially if the alternative spatial lags are highly correlated. Instead of solely focusing on the specification of W, researchers should reflect more thoroughly on temporal dynamics, common shocks, and unit heterogeneity when building their spatial econometric models since these specification issues can distort the estimates of the spatial effect.