Julian Schuessler

When using instrumental variables, researchers often assume that causal effects are only identified conditional on covariates. We show that the role of these covariates in applied research is often unclear and that there exists confusion regarding their ability to mitigate violations of the exclusion restriction. We explain when and how existing adjustment strategies may lead to bias. We then discuss assumptions that are sufficient to identify various treatment effects, some of which are new, when the exclusion restriction only holds conditionally. In general, these assumptions are highly restrictive, albeit they sometimes are testable. We also show that other existing tests are generally misleading. Then, we introduce an alternative sensitivity analysis that uses information on variables influenced by the instrument to gauge the effect of potential violations of the exclusion restriction. We illustrate it in two replications of existing analyses and summarize our results in easy-to-understand guidelines.

A series of papers uses administrative data on school students' grades to assess whether teachers discriminate against certain demographic groups. Often, standardized test grades are subtracted from teacher grades and then regressed on student-level variables. However, it is unclear under what circumstances such an estimation strategy is valid. We conceptualize teacher bias as a direct causal effect of student-level attributes on teacher grades, fixing student ability. Standardized tests merely proxy for student ability; additionally, there may be confounders of ability and teacher grade. Accordingly, teacher bias is nonparametrically unidentified. However, we suggest substantive and parametric assumptions that ensure identification using difference-in-grades estimators. Estimators based on regression control for test grades are shown to be inconsistent even under these strong assumptions. We then develop a parametric sensitivity analysis that allows researchers to investigates the consequences of departures from critical assumptions. We illustrate our methodology using administrative data from Denmark.

Post-instrument covariates are often included as controls in IV analyses to address a violation of the exclusion restriction. However, we show that such analyses are subject to biases unless strong assumptions hold. Using linear constant-effects models, we present asymptotic bias formulas for three estimators (with and without measurement error): IV with post-instrument covariates, IV without post-instrument covariates, and OLS. In large samples and when the model provides a reasonable approximation, these formulas sometimes allow the analyst to bracket the parameter of interest with two estimators and allow the analyst to choose the estimator with the least asymptotic bias. We illustrate these points with a discussion of Acemoglu, Johnson, and Robinson (2001).

Differentiated integration is often considered a solution to gridlock in the European Union. However, questions remain concerning its perceived legitimacy among the public. While research shows that most citizens are not, in principle, opposed to differentiated integration – although support varies across different differentiated integration models and different country contexts – we still know little about the role institutional design plays in citizens’ evaluations of differentiated integration. This article inspects how citizens evaluate different hypothetical differentiated integration arrangements, with varying decision-making procedures, using a conjoint experiment. We ask whether institutional arrangements can overcome citizens’ preference heterogeneity over differentiated integration, and thereby foster the legitimacy of a differentiated European Union. We find that while a majority of citizens care about the inclusiveness of differentiated integration arrangements, they also support limiting the number of veto points. Our analysis also reveals noteworthy differences across citizens with pro- and anti-European Union attitudes in the perceived fairness of differentiated integration arrangements.

Directed acyclic graphs (DAGs) are an increasingly popular tool to inform causal inferences in observational research. We demonstrate how DAGs can be used to encode and communicate theoretical assumptions about nonprobability samples and survey nonresponse, determine whether typical population parameters of interest to survey researchers can be identified from a sample, and support the choice of adjustment strategies. Following an introduction to basic concepts in graph and probability theory, we discuss sources of bias and assumptions for eliminating it in selection scenarios familiar from the missing data literature. We then introduce and analyze graphical representations of the multiple selection stages in the survey data collection process, in line with the Total Survey Error approach. Finally, we identify areas for future survey methodology research that can benefit from advances in causal graph theory.

There is a long-standing interest in how the visual appearance of politicians predict their success. Usually, the scope of such studies is limited by the need for human-rated facial features. We instead fine-tune pre-trained image classification models based on convolutional neural networks to predict facial features of multiple thousand Danish politicians. Attractiveness and trustworthiness scores correlate positively and robustly with both ballot paper placement (proxying for intra-party success) and the number of votes gained in local and national elections, while dominance scores correlate inconsistently. Effect sizes are at times substantial. We find no moderation by politician gender or election type. However, dominance scores correlate significantly with outcomes for conservative politicians. We discuss possible causal mechanisms behind our results.

Research in social psychology has long argued that exposure to objectifying portrayals of women can lead to increasingly misogynist attitudes and behavior. We argue that such images can also impact on gendered policy attitudes. We suggest that objectifying images prime sexist attitudes and reduce perceptions of women’s agency, warmth, and competence. We argue that this may translate into decreased support for reproductive rights and other gender-salient policies. Furthermore, these effects may vary by the gender of those exposed to these images. In two survey experiments with brief exposures to objectifying images, we find mixed support for these predictions. Although we find some negative effects as predicted, we also find positive effects of objectification among women in the sample that are suggestive of a backlash effect. We discuss potential explanations for this heterogeneity. Overall, our results suggest interesting avenues to further explore the effects of objectification on political outcomes.

Previous research emphasizes that individual economic status does not significantly influence support for redistribution within the European Union (EU). Instead, identity factors are often posited as the main causes. We study the interaction of these variables and synthesize various theories that all predict that heightened European identification leads to a weaker influence of economic status. In a large original survey fielded in 12 countries, we find that respondents' income and perceived relative position correlate negatively with their redistribution preferences, both on the national and the EU level, as predicted by economic accounts. We also replicate findings on the positive effect of identity variables and find some evidence for the predicted interaction. However, randomized information treatments aimed at altering perceptions of an individual's or their member state's relative economic position fail to impact on preferences and do not interact with identity variables. Overall, our findings point toward a possible but quantitatively very limited role of economic status and its interaction with identity in understanding EU redistribution preferences.

We suggest that discriminatory practices may vary significantly across decision-makers, which allows for deeper insights into the mechanisms behind discrimination. We study this in the context of biased grading in schools. We develop a theory of teacher biases driven by heuristic beliefs stemming from concrete classroom experiences. Because teachers may also care about grade equality, such a mechanism can lead to either inequality-reinforcing or compensating biases in grading. Based on large-scale administrative data on Danish students, we find strong evidence for highly heterogeneous teacher biases---up to 45\% of teachers exhibit a bias that is of the opposite sign as the average bias. Furthermore, there is a robust and substantively large compensation effect. Teachers that experienced a visible demographic group (defined by gender or migration background) academically under-performing relative to a reference group show more positive bias towards that group than teachers where the same group over-performed. We find little evidence for alternative explanations of bias. To fully grasp discrimination, we must go beyond averages and consider the wide variety of biases shaped by individual experiences.

This article maps and investigates public support for different types of differentiated integration (DI) in the European Union. We examine citizens’ preferences for DI using novel survey data from eight EU member states. The data reveals substantive differences in support for different types of DI. Factor analyses reveal two dimensions that seem to structure citizens’ evaluations of DI. The first dimension relates to the effect of DI on the European integration project, the second concerns the safeguarding of national autonomy. Citizens’ attitudes on this second dimension vary substantively across countries. General EU support is the most important correlate of DI support, correlating positively with the first and negatively with the second dimension. Our results underline that while citizens generally care about the fairness of DI, balancing out their different concerns can be a challenging political task.

While billions have been vaccinated against COVID-19, unvaccinated citizens remain a challenge to public health given their higher likelihood of passing on the virus. One way for governments to reduce this concern is to enact more restrictive rules and regulations for the unvaccinated citizens in order to incentivize them to become vaccinated and/or reduce their spread of the virus. However, such rule differentiation conflicts with liberal principles of equal treatment, thereby raising a trade-off between material (public health) and principled concerns. To gain legitimacy in trading off these difficult concerns, governments are likely to look to preferences in the general population. We therefore analyze to what extent unequal treatment of the unvaccinated in terms of differentiation of various rules and regulations finds support among the general public. In a pre-registered survey experiment, we investigate public support for various COVID-19 regulations (e.g., test fees, isolation pay, and hospital prioritization). In the experiment, we randomly assign respondents to evaluate regulations that either (i) apply to adults in general or (ii) only to those adults who deliberately have chosen not to be vaccinated. This design provides a valid means to assess support for unequal treatment of the unvaccinated by minimizing various concerns relating to survey responding. Furthermore, we examine how these preferences vary by individual vaccination status, trust in institutions, as well as over-time changes in severity of the pandemic. We find significantly (both statistically and substantively) higher support for restrictive policies when targeted exclusively toward the unvaccinated, which we interpret as support for unequal treatment of this group. We also uncover strong polarization in these preferences between the vaccinated and the unvaccinated, but a much more limited role for trust and severity of the pandemic.

Research on differentiated integration (DI) in the European Union has burgeoned in recent years. However, we still know little about citizens’ attitudes towards the phenomenon. In this article, we argue that at the level of individual citizens, liberal economic values increase support for DI. Stronger preferences for equality, in contrast, make opposition to the concept more likely. Similarly, concerns about discriminatory differentiation at the member state level lead citizens to oppose DI. We test the theoretical claims by analysing survey data on citizens’ attitudes towards a ‘multi-speed Europe’. Supporters of DI, indeed, are marked by liberal economic attitudes. In contrast to general EU support, we do not find robust correlations with socio-demographic variables. Moreover, the data reveal striking differences amongst macro-regions: support for DI has become much lower in Southern European states. We attribute this opposition to negative repercussions of the Eurozone crisis.

Conjoint experiments aiming to estimate average marginal component effects and related quantities have become a standard tool for social scientists. However, existing solutions for power analyses to find appropriate sample sizes for such studies have various shortcomings and accordingly, explicit sample size planning is rare. Based on recent advances in statistical inference for factorial experiments, we derive simple yet generally applicable formulae to calculate power and minimum required sample sizes for testing average marginal component effects (AMCEs), conditional AMCEs, as well as interaction effects in forced-choice conjoint experiments. The only input needed are expected effect sizes. Our approach only assumes random sampling of individuals or randomization of profiles and avoids any parametric assumption. Furthermore, we show that clustering standard errors on individuals is not necessary and does not affect power. Our results caution against designing conjoint experiments with small sample sizes, especially for detecting heterogeneity and interactions. We provide an R package that implements our approach.

The European Union targets up to a quarter of its budget towards underdeveloped regions. Do these investments have an impact on citizens' attitudes towards the EU? The previous literature on this question is scarce and inconclusive. I use a regression- discontinuity design to tackle this question. The analysis is based on a large dataset of geocoded individual-level survey responses from every member state that spans more than twenty years. Point estimates for the effect of the funds on public opinion are relatively small and statistically insignificant. Large effects can be ruled out. At the same time, I show that pre-existing attitudes towards European integration correlate with the EU's allocation decisions. Finally, I show that the funds do not have an impact on EU-related attitudes due to informational problems and explore whether the EU's activities may be misattributed to national political actors.