Online Seminar Talk

Polarized Social Media Users or Polarized Ads? Mechanisms That Shape the Sample Composition in Social Media Recruited Surveys

Zaza Zindel, Simon Kühne
Laboratory of Digital and Computational Demography, February 16, 2022


A growing number of research projects are turning to social networks such as Facebook to recruit (rare) populations for online survey participation. These social media platforms are based on an advertising revenue model that researchers can leverage by purchasing advertisements to recruit survey participants. However, researchers have limited control over the algorithms that allocate ads to users. Moreover, the goals of survey researchers (e.g., obtain representative survey data) do not necessarily match those of social media platforms or advertisement management systems (to maximize interaction and profit). Consequently, ad allocation algorithms may lead to polarized response patterns that favor users with very strong opinions on certain survey topics. Thus far, survey research lacks insights into 1) how the advertising algorithms allocate ads to users and 2) how different ad designs affect the composition of survey participants. We plan to contrast different ad images and campaign options for recruiting survey respondents via Facebook. In doing so, we want to investigate whether polarizing stimuli not only affect ad interactions but are also reflected in survey data. This project should thus provide valuable insights into data quality hurdles such as sampling bias when using social media for survey recruitment.


Simon Kühne is professor of Applied Social Data Science at Bielefeld University. His research focuses on survey methodology, new data sources for social science (esp. social media), and social inequality. He is an expert in survey sampling and weighting. Prior to his position at Bielefeld University, he worked at the Socio-Economic Panel Study (SOEP, DIW Berlin), where he was involved in the sampling and weighting of the latest migrant/refugee samples.

Zaza Zindel is a PhD candidate in sociology at Bielefeld University. Her research interests include survey research and methodology, social media research, discrimination, and social inequality. Since 2020, she is working on the project “Gender Identity and Sexual Orientation Diversity in Focus: Participation and Diversity in Human Beings (SOEP-GeSMin)”, where she is responsible for designing and implementing the social media-generated LGBTQI+ samples.

The Max Planck Institute for Demographic Research (MPIDR) in Rostock is one of the leading demographic research centers in the world. It's part of the Max Planck Society, the internationally renowned German research society.