Recruiting in an Internet panel using respondent driven sampling
*Matthias Schonlau, University of Waterloo
Keywords: Respondent Driven Sampling, Internet Panel, Hispanic, American Life Panel, web survey
Respondent driven sampling (RDS) is a sampling technique typically employed for hard-to-reach populations (e.g. HIV populations, drug users, men who have sex with men, jazz musicians, immigrants). Briefly, initial seed respondents recruit additional respondents from their network of friends. The recruiting process repeats iteratively, thereby forming long referral chains. It is crucial to obtain estimates of respondents’ network size (e.g. number of friends with the characteristic of interest) and to know who recruited whom. RDS shares some similarities with snowball sampling, but the theoretical foundation for inference using RDS samples is much stronger. We used RDS as a recruiting strategy for the American Life panel (ALP). Because RDS is typically conducted face-to-face rather than on the web, there is no guidance on how to ensure RDS assumptions are met for a web-based sampling effort. In particular, it is unclear how to best elicit the number of friends in the target population and how to facilitate the recruiting process to avoid referral chains dying out. We present results from two experiments on respondents on the American Life Panel (ALP), and from an implementation in the ALP based on the experiments. While the experiments identified a strategy for sustainable wave-to-wave recruiting, results from the ongoing implementation are mixed.