How does broadband supply affect participation in panel surveys? Using geospatial broadband data at the district level to analyze mode choice and panel attrition

Abstract

Good broadband supply is crucial for different aspects of participation behavior in web surveys. In this study, we combined geospatial broadband data and the survey data of 16 waves of the mixed-mode GESIS Panel to investigate mode choice and panel attrition. Since small-scale geospatial data is often unavailable or difficult to access for research purposes, we report here results of a feasibility study investigating whether publicly available broadband data at the district level are sufficient to draw meaningful conclusions about participation behavior. In the first part of the analysis, we investigated the effects of broadband supply on mode choice. The results showed a positive effect of the broadband category with the highest coverage rate on choosing the online mode. We also found effects for internet familiarity, age, and education. In the second part, we investigated the effects of broadband supply on panel attrition within the 16 survey waves. We did not find significant effects of broadband supply on panel attrition. However, we found effects for evaluation of duration, measured duration, and age. Overall, we conclude that the granularity of the geospatial data on broadband supply is not ideal. A detailed discussion on the granularity of the broadband data in the context of the general availability of fine-grained geospatial data is provided at the end of this study.