Book Chapter in "Applied Panel Data Analysis for Economic and Social Surveys" |
Together with Hans-Jürgen Andress and Katrin Golsch. The book deals with advanced methods for the analysis of panel data. We present models for continous and categorical dependent variables. In my contribution to this book I demonstrate how to create and manange panel data sets. Furthermore, I discuss practical problems that occur when we analyze real survey data. In particular, I discuss issues of balanced and unbalanced as well as weighted and unweighted analyses. The book has been published in January 2013 at Springer. Find the book's webpage here.
Do Parties "Playing the Race Card" Undermine Natives' Support for Redistribution? Evidence from Europe |
The paper is published in Comparative Political Studies, 47(4), pp. 519 - 549.
DOI:
10.1177/0010414013488542.
Together with Dennis Spies. We investigate the influence of anti-immigrant sentiments on welfare state support. In particular, we are intersted in the moderation of this relationship by the party discourse. We use data from the comparative party manifesto project to operationalize the political discourse with regard to migration and investigate how it moderates the relationship between attitudes toward migration and welfare state solidarity. Our results show that anti-immigrant attitudes can undermine the support for redistribution. The results indicate that the party discourse, indeed, moderates this relationship. The more parties accentuate migration-critical issues, the stronger the negative relationship between anti-immigrant sentiments and welfare support. Find the online appendix here.
Social Presence in Online Surveys |
The paper is published in Methoden, Daten, Analysen, 7(3), pp. 319 - 432.
DOI: 10.12758/mda.2013.019.
Together with Katharina Hörstermann. Using an experimental design, we investigate the positive and negative effects of simulated social presence in online surveys. We expected social presence to reduce non-response rates but to increase the share of social desirable answers. We, furthermore, investigate whether gender-of-interviewer effects are present in online-surveys if social presence is simulated with photos. In addition to gender, we varied the attractiveness of the interviewers shown on the photos. The results show that simulated social presence does not help in reducing non-response. Furthermore, the results demonstrate that social presence in general, and the gender and attractiveness of the simulated interviewer in particular, could cause systematic differences in responses. We conclude that simulated social presence in online surveys does not invoke positive effects on the survey quality but might have negative impacts in the form of interviewer effects.
Economic Inequality and Public Demand for Redistribution: Combining Cross-sectional and Longitudinal Evidence |
The paper is published in Socio-economic Review, 14(1), pp. 119 - 140.
DOI: 10.1093/ser/mwu030.
Is economic inequality related to support for redistribution? The median-voter hypothesis claims this relationship to exist. Empirical evidence, however, is still rather ambiguous. In this paper I combine cross-sectional and longitudinal methods to test whether inequality increases the demand for redistribution. From a pure economic perspective this is likely to be true. From a sociological perspective inequality might also reflect redistributive ethics or a taste for equality. By comparing cross-sectional and longitudinal effects I evaluate the value of both types of explanations. The results suggest that increasing inequality, indeed, leads to higher demand for redistribution. This relationship can be identified only within countries, while it does not hold between countries. Hence, there must be something unobserved between countries that is related to demand for redistribution and inequality. The candidates for this unobserved variable are "worlds of welfare attitudes". I employ a variety of welfare regime classification and test whether they capture the unobserved differences in demand for redistribution. The results demonstrate the limited value of the welfare regime approach with regard to the explanation of attitudes.
The Non-uniqueness Property of the Intrinsic Estimator in APC Models |
The paper is published in Demography, 52(1), pp. 315 - 327.
DOI: 10.1007/s13524-014-0360-3.
Together with Ben Pelzer, Manfred te Grotenhuis and Rob Eisinga. The Intrinsic Estimator has recently been proposed as a solution to the APC problem. Using formal proofs and empirical examples, we demonstrate the previously not recognized non-uniqueness of the Intrinsic Estimator. Thereby, our results pose a challenge to the usefulness of the Intrinsic Estimator. The paper has been submitted in September 2013 and was accepted for publication in Demography in September, 2014. It was published online in December 2014.
A follow-up paper has also been published in Demography.
(online first: DOI: 10.1007/s13524-016-0476-8).
Multilevel Models for Comparative Longitudinal Survey Data |
Published in European Sociological Review, 32(1), pp. 23 - 38.
DOI: 10.1093/esr/jcv090.
Together with Malcolm Fairbrother. Multilevel models have become increasingly popular in the social sciences. In recent years, scholars have started to fit multilevel models to pooled cross-sectional survey data. Such data could potentially provide important information about social change and causal relationships between variables. On the other hand, such data requires to specify more complex error structures. We observe that many scholars make significant mistakes in specifying their multilevel models. We demonstrate that these misspecifications can result in seriously biased estimates and standard errors and propose how to model such data correctly. The project started in September, 2013. The paper was submitted in April, 2014 and accepted for publication in the European Sociological Review in April 2015.
Immigration and Welfare Support in Germany |
The paper is published in American Sociological Review, 81(2), pp. 242 - 261.
DOI: 10.1177/0003122416633140.
Together with Dennis Spies. In this project we are interested in the effect of migration on natives' support for welfare. We use regional data of German communities to investigate whether an increase in the migrant population leads to a decline in natives' support for welfare. This hypothesis has been established but is mostly tested on the level of countries. In our study we use time series data for small regional units in order to provide a solid test of the hypothesis. The results suggest that migration, indeed, leads to less support for welfare policies by the native population.
Migration, Migrant Integration and Support for Social Spending: The Case of Switzerland |
Published in Journal of European Social Policy, , 26(1), pp. 32 - 47.
DOI: 10.1177/0958928715612170.
Together with Dennis Spies. An extensive body of scholarship has claimed that the relationship between migration and the welfare state is a potentially troublesome one, because the native population might be concerned about the fiscal, economic and cultural threats this poses. At the same time, studies have argued that "migrants differ", not only in their actual numbers but also in their similarities or differences compared with the native population. Taking these differences into account, we analyse the effect of the integration of migrants for natives' support for welfare. In detail, we test for the possibility that the integration of migrants might have a direct impact on the economic and cultural difficulties which natives associate with migration and in this way will have an indirect effect on their support for social spending. Our results show that the objective integration of migrants has only limited relevance for the relationship between migration and welfare support and point to the need to focus on subjectively perceived migration and integration-related attitudes of natives. The paper has been accepted for publication in December, 2014.
Institutions, culture and migrants' preference for state-provided welfare. Longitudinal evidence from Germany |
Forthcoming in European Journal of Social Policy. Download the manuscript here!
Together with Romana Careja. In this project we investigate the development of welfare attitudes of migrants. Using data from the German Socio-Economic Panel Study we investigate migrants' welfare attitudes over time. We test the hypothesis that migrants' attitudes toward welfare approach the welfare attitudes of natives. This hypothesis is based on the idea of a socializing effect of welfare state insititutions. It challanges the claim of some authors who argue that differences between natives' and migrants' welfare attitudes are due to cultural differences that remain stable over time. Our findings show that natives' and migrants' attitudes towards state provided welfare converge over time. The paper has been submitted in February, 2015, received an R&R in April and got accepted in November, 2015.