Skip to main content

Estimation from Network-Based Respondent-Driven Sampling

Respondent-Driven Sampling employs a variant of a link-tracing network sampling strategy to collect data from hard-to-reach populations. By tracing the links in the underlying social network, the process exploits the social structure to expand the sample and reduce its dependence on the initial (convenience) sample.

Current estimation focuses on estimating population averages in the hard-to-reach population. These estimates are based on strong assumptions allowing the sample to be treated as a probability sample. In particular, the current estimator assumes a with-replacement sample or small sample fraction, while in practice samples are without-replacement, and often include a large fraction of the population. We demonstrate the sensitivity of the current estimator to violations of these assumptions. We then introduce two new estimators which allow for the without-replacement structure of the sample, and demonstrate their superior performance, particularly in cases where the sample fraction is large.