SOEP Research: Survey Methodology and Data Science

The Beyondpareto Command for Optimal Extreme-Value Index Estimation

In this article, we introduce the command beyondpareto, which estimates the extreme-value index for distributions that are Pareto-like, that is, whose upper tails are regularly varying and eventually become Pareto. The estimation is based on rank-size regressions, and the threshold value for the upper-order statistics included in the final regression is determined optimally by minimizing the asymptotic ...

Introduction to Lab²: A MetaLab for Better Science

The replicability crisis highlights the need for more transparent and robust research practices. Lab² (Lab-Square) is a collaborative initiative using meta-scientific approaches to improve research credibility. Through large-scale projects like Many-Labs, Many-Designs, and Many-Analysts, we assess the generalizability of findings across populations, methods, and analytical approaches. Lab² also...

Regional Decline, Satisfaction with Democracy and Support for Right-Wing Populist Parties. Evidence from Germany (with Larissa Deppisch)

A consistent finding in research on the success of right-wing populist parties is that they gain support in regions that are peripheralized. In such regions, the decline of manufacturing jobs, public services, and infrastructure is thought to lead to growing frustration with democratic institutions and mainstream political parties, providing opportunities for right-wing populist parties to...

Learn to use the SOEP over lunch

The German Socio-Economic Panel Study is a representative panel study for the German population, collecting data on a broad variety of topics of everyday life, including general well-being, household composition, educational aspirations and educational status, income and occupational biographies, leisure time activities, housing, health, political orientation and more. With its long running panel...

Who Benefits from Place-based Policies? Evidence from Matched Employer-Employee Data (with Philipp Grunau, Florian Hoffmann, and Thomas Lemieux)

We study the wage and employment effects of a German place-based policy using a research design that exploits conditionally exogenous EU-wide rules governing the program parameters at the regional level. The place-based program subsidizes investments to create jobs with a subsidy rate that varies across labor market regions. The analysis uses matched data on the universe of establishments and...

The #ManyDaughters Study is Officially Kicking Off!

We are happy to announce the launch of the #ManyDaughters Study, an international research initiative exploring how having daughters influences behavior, preferences, and attitudes. Researchers from all fields of social sciences are invited to participate in this collaborative project, which will use data from the German Socio-Economic Panel (SOEP) to test four key hypotheses on the impact of daughters ...

Dealing with Censored Earnings in Register Data

Earnings are often top-coded (right-censored) in administrative registers. The censoring threshold in the case of Germany is the limit value for social security contributions, leading to a substantial fraction of censoring: For example, about 12%of male workers inWest Germany are affected, rising to above 30% for highly educated prime-aged workers. This missing right tail of the earnings distribution ...

SOEP Annual Report 2024

Do talk money – Reducing income nonresponse in surveys (with Katharina Allinger)

Item nonresponse is a common issue in surveys. We implement an experiment to reduce nonresponse to income questions in an international household survey, looking at four different countries. Survey respondents are asked to report their exact household income. We randomize those who refuse to answer into two groups. In a follow-up question, the control group is asked to choose their income from a...

Validating an Index of Selection Bias for Proportions in Non-probability Samples

Fast online surveys without sampling frames are becoming increasingly important in survey research. Their recruitment methods result in non-probability samples. As the mechanism of data generation is always unknown in such samples, the problem of non-ignorability arises making vgeneralisation of calculated statistics to the population of interest highly questionable. Sensitivity analyses provide a ...