Respondent Driven Sampling

http://www.respondentdrivensampling.org/

http://www.migrationinformation.org/USfocus/display.cfm?ID=441

 

Respondent Driven Sampling (RDS) is a technique used by sociologists and other researchers to study hidden populations.  It is especially hard to identify and reach such populations in order to conduct research on them because data assigning people into these demographics is not readily available.  Some examples of hidden populations include second generation immigrants, injection drug users, sex workers, and other groups.  RDS uses the concept of graph theory and the related characteristics that define social networks to draw conclusions about a given population.  The success of RDS in representing a target population is based on the connectivity of social networks and homophily within the social network of the chosen hidden population.    

The process works by identifying a small group of seeds that meet the criteria of the group to be studied.  The seeds then complete the survey and refer a specific number of their friends to complete it as well.  The process continues until the desired sampling size has been attained.  In the process a directed graph is formed with waves of referral that shows the distance between members of the group.  The network created is then structurally analyzed and statistically inferences are made from the pattern that was formed. 

This method of chain-referral has been criticized for its inability to provide accurate statistical data on the sampling; however, the above article suggests that if inferences are made from the social network encompassing the population instead of directly from the population itself unbiased estimation techniques may exist.  There is evidence supporting the validity of RDS that supports the idea that people will recruit their friends randomly when participating in the study.  RDS has been used to combat the problem of acquiring willing interviewees since it is common that members of hidden populations are engaging in illegal activity.  When people are referred by a friend with an established notion of trust they are more likely to participate in the study.  RDS relies on the insular nature of hidden populations to draw conclusions about their characteristics and has been useful in obtaining otherwise unknown data.

 

Posted in Topics: Education

Responses are currently closed, but you can trackback from your own site.

Comments are closed.



* You can follow any responses to this entry through the RSS 2.0 feed.