This paper describes the use of respondent-driven sampling (RDS) to identif
y and survey jazz musicians in four U.S. cities-Detroit, New Orleans, New Y
ork and San Francisco. RDS has been used previously to sample injection dru
g users as part of a study of HIV prevention education and services. Throug
h a mechanism of long referral chains and modest incentives, RDS will help
to identify the social structure of the jazz musicians' community in the fo
ur study cities and their patterns of affiliation. It will also help to doc
ument everyday practice and to identify areas of greatest need for living j
azz musicians. (c) 2001 Elsevier Science B.V. All rights reserved.