Respondent-driven sampling (RDS): The form of snowball sampling which starts with the recruitment of the small number of people in the target population to serve as the seeds. After participating the seeds are asked to recruit other people they know in target population. The sampling continues in this manner with the current sample members recruiting the next wave of the sample members until the desired sample size is achived. By using the mathematical model which weights the sample to compensate for the fact that the sample was collected in a non-random manner, the data provided by such a sampling scheme can be used to give asymptotically unbiased estimates about target population. An instance of the use of this approach is the estimation of the drug user's in New York who have HIV.