Surveying Panel Participants’ Network Members: Integration of Egocentric Data Collection and Respondent-Driven Sampling

Individuals are embedded in social contexts: although many large-scale, cross-thematic social science surveys already attempt to do justice to this fact, so far research has only been partially successful.

The aim of this study is to supplement and improve the existing data collection strategies - survey designs for multiple actors (MA) and questionnaire modules for egocentric networks (ECN) - with Respondent-Driven Sampling (RDS). RDS is a network sampling strategy in which survey participants act as “seeds” who recruit their network members to participate in the survey. They, in turn, recruit network members for the survey and so on, so that reference chains are built up. Using a link tracing process, each recruited person can be traced back to the original source via their recruiters. Questions that can be answered with this data include a) whether and to what extent broader network bubbles exist (e.g. familial, regional, political); b) how strong, transitive, homogeneous these network bubbles are; and c) how far from the seeds one has to go along the reference chain before their network bubbles burst and become heterogeneous.

The project aims to achieve three objectives: 1) to investigate practical innovation with regard to the success of RDS by effectively combining it with ECN modules, i.e. by establishing long reference chains; 2) to develop and apply strategies to assess and reduce bias in the resulting data; and 3) to close existing research gaps on social networks with regard to social cohesion. These goals will be achieved by expanding the data collection wave of the German Social Cohesion Panel (SCP) in 2025 to include an RDS procedure and analyzing the resulting network data together with the panel data.