Assessing Feasibility of Respondent-Driven Sampling Using Pilot Data with an Application to Older Lesbian, Gay, and Bisexual Adults

• Activity Number: 484
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Government Statistics Section
• Author(s): Maryclare Griffin* and Elena Erosheva and Karen Fredriksen-Goldsen
• Companies: University of Washington and University of Washington and University of Washington
• Keywords: respondent-driven sampling ; link-tracing ; network sampling ; networks ; hard to reach populations

Abstract

Respondent driven sampling (RDS) is a method for estimating features of hard to reach populations. However, RDS is feasible when social networks of hard to reach populations are well connected. Although there is interest in assessing the feasibility of RDS using pilot data, methods for doing so are not well developed. Therefore, we introduce an approach for assessing the feasibility of RDS using sampled, aggregated relational data (ARD), which measures the number of connections sampled individuals (egos) have to other individuals (alters) by alter type. We develop a method for estimating network connectivity from sampled ARD and assessing RDS feasibility given estimates of network connectivity via simulation of RDS on synthetic networks. We estimate measures of network connectivity from ARD, simulate synthetic population networks consistent with these estimated measures, and simulate RDS on the synthetic population networks. We allow for cases where more is known about egos than alters and cases where population sizes are partially known. We develop this approach in general and illustrate its use for RDS feasibility in populations of older lesbian, gay and bisexual adults.