Forschung SOEP: Survey-Methodologie und Data Science

Exploring Integration and Migration Dynamics: The Research Potentials of a Large-Scale Longitudinal Household Study of Refugees in Germany

Forced migration has intensified in the 21st century, driven by conflicts, persecution, and political instability in regions such as the Middle East, Afghanistan, the Horn of Africa, South-East Asia, Latin America and, most recently, Ukraine. Germany has become a primary destination for refugees within the European Union and one of the largest among the OECD countries. The IAB-BAMF-SOEP Refugee Survey, ...

Education Bias in Probability-Based Surveys in Germany: Evidence and Possible Solutions

This paper outlines two studies on education bias in German probability-based surveys. Study 1 reviews data from 67 surveys across 19 survey programs conducted in Germany from 2000 to 2023. We found a consistent underrepresentation of individuals with a low level of formal education. We also found that the transition to self-administered modes due to rising survey costs may exacerbate this bias in ...

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 ...

Composition of Core Modules and Item Allocation in Split Questionnaire Designs: Impact on Estimates from Imputed Data

An increasing number of social science surveys use split questionnaire designs to reduce questionnaire length, presenting only a subset of several questionnaire modules to each respondent while leaving out others. This approach results in large amounts of planned missing data that necessitates imputation. Research shows that imputation is most effective when each module covers various topics. Yet, ...

Measurement error models on anonymized georeferenced data

Georeferenced data are often anonymized for data protection reasons. This is done either by aggregating the data into larger spatial units (e.g., higher-level administrative units or grids with larger cell sizes) or by using stochastic methods to deliberately overlay the original coordinates. These methods significantly distort the data and associated variables, making further modeling steps...

Messfehlermodelle für anonymisierte, georeferenzierte Daten

Der englische Originaltitel des Seminars lautet: “Measurement error models on anonymized georeferenced data”. Die Präsentation findet auf Englisch statt. Eine kurze Zusammenfassung zum Vortrag ist nur auf der englischen Veranstaltungsseite verfügbar!

Methodische Herausforderungen bei kohortenübergreifendem Vergleich von Bildungsleistungen

Methodological Challenges in Cross-Cohort Comparisons of Educational Achievement

Filling in the Blanks: Augmenting Survey Data Imputation with an External Prior

Multiple imputation of missing values in survey data analysis is a state-of-the-art technique. Typically, methods like multivariate imputation by chained equations (mice, van Buuren 2018) are employed, replacing missing values on a variable-by-variable basis. Generally, the information used for imputation comes from the survey dataset being analysed. Valid analysis results are achieved when the...

Lücken füllen: Ergänzung der Imputation von Umfragedaten durch eine externe Prior-Verteilung

Der englische Originaltitel des Seminars lautet: “Filling in the Blanks: Augmenting Survey Data Imputation with an External Prior”. Die Präsentation findet auf Englisch statt. Eine kurze Zusammenfassung zum Vortrag ist nur auf der englischen Veranstaltungsseite verfügbar!