Blanka Szeitl

Hungarian Academy of Sciences

Received Sylff fellowship in 2023
Academic supervisor: Tamás Rudas
Current affiliation: PhD student

Blanka Szeitl graduated in 2015 with a master's degree in Survey Statistics from the Faculty of Social Sciences of ELTE. Since 2017 she has been a PhD student in Applied Mathematics at the University of Szeged. Her PhD research topic is the development of stochastic survey sampling techniques and the investigation of new quantitative methodological directions. In addition, she is also working on the analysis of data collection biases that may arise from non-response and new data collection platforms. She is involved in projects related to the European Social Survey (ESS) at the Centre for Social Sciences (ELKH Hungarian Academy of Sciences Centre of Excellence).

Survey research has a central role in empirical social research. The methods of data collection and their accuracy have been greatly influenced by technological developments and changes in the social environment in recent decades. Empirical social research draws conclusions by analyzing objective data, so the reliability and accuracy of the data are crucial. The aim of the research is to develop and test post-stratification procedures to handle bias due to non-response and measurement uncertainty in estimates from face-to-face and online samples. Although there is international literature investigating these types of weights, they are not generally applicable in all countries, as differences in non-response patterns, measurement uncertainties, and internet coverage are country-specific. In my work, I construct and test post-stratification weights based on national characteristics by developing my own simulation environment. My current proposal also builds on my previous results and can be considered as foundational research, as in the future these weighting procedures could be tested on real data sets and would lead step by step to more accurate and ready-to-use methods.


To contact this fellow, email the Sylff Association at sylff[a] (replace [a] with @).