In the field of survey statistics, we investigate different aspects of the quality of surveys. Sample design, in particular sample size and sample composition, have a direct impact on the representativeness of survey results. Unit nonresponse and item nonresponse pose further challenges to data quality. The research activities in the field of survey statistics therefore focus on problems and solutions associated with random sampling. These are also investigated in several third-party funded projects.
Research topics are in particular:
- Drawing procedures for complex sample designs
- Nonresponse (bias) analysis
- Weighting procedures for survey designs
- Imputation of missing values
- Variance estimation under complex sample designs and imputation of missing values
- Application of machine learning methods in survey statistics
-
Data harmonization
Tools and topic-specific harmonized data for your research project
-
Harmonisation
Consulting on the harmonization of content elements and social-structural characteristics
-
Harmonizing and synthesizing partnership histories
A tool for harmonization and synthesis of couple biographical data.
-
Harmonizing substantive instruments
Beratung zur Harmonisierung von Instrumenten zur Messung latenter Konstrukte
-
Imputation of Missing Values
We perform imputation of missing values for your complete data set or even selected variables
- Bruch, Christian, and Barbara Felderer. 2022. "Prior Choice for the Variance Parameter in the Multilevel and Poststratification Approach for Highly Selective Data: A Monte Carlo Simulation Study." Austrian Journal of Statistics 51 (4): 76-95. doi: https://doi.org/10.17713/ajs.v51i4.1361.
- Bruch, Christian, and Barbara Felderer. 2022. "Applying Multilevel Regression Weighting When Only Population Margins Are Available." Communications in Statistics - Simulation and Computation online first. doi: https://doi.org/10.1080/03610918.2021.1988642.
- Quandt, Markus, and Antonia May. 2019. "Classifying a Large Number of Countries by their Populations’ Shares of Ethnic and Civic Types of National Identification." European Survey Research Association conferences, University of Zagreb, Zagreb.
- Quandt, Markus, and Antonia May. 2019. "Using Latent Class Analysis for testing the equivalence of multi-item scales across data from different survey programs. The example of National Identity types." Conference of the ‘Survey Data Recycling Project’: Building Multi-Source Databases for Comparative Analyses, Institute of Philosophy and Sociology of the Polish Academy of Sciences Warsaw, Poland, 2019-12-16.
- Bluemke, Matthias. 2018. "Bachelor-Thesis-Gutachten (Uni Heidelberg)." Anzahl: 1.