NCJ Number
215537
Journal
Journal of Drug Issues Volume: 36 Issue: 2 Dated: Spring 2006 Pages: 417-440
Date Published
2006
Length
24 pages
Annotation
Based on data obtained in the evaluation of the nationwide drug prevention demonstration program called Fighting Back, this study considered the feasibility of using the network scale-up method to estimate heroin use.
Abstract
The study shows that the network scale-up method is a promising addition to the methods currently available for estimating the size of difficult-to-count populations such as heroin users in different communities. With just six subpopulations used as predictors, the authors estimated the prevalence of heroin use in 14 mid-sized U.S. cities. The network scale-up method, which has been used since 1986, involves asking a statistically representative set of respondents how many people they know in a variety of subpopulations. "Know" is defined for respondents as their having had contact within the last 2 years. In the current study, the subpopulation about which respondents were asked was individuals whom they knew who used heroin. The method assumes that, other things being equal, the probability that any members of the respondent's network belonging to a particular subpopulation is the fraction of the larger population constituted by that subpopulation. The data reported in the current study were obtained from the second wave of a telephone survey conducted in the spring of 1997 as part of the evaluation of a nationwide, community-based drug and alcohol prevention program (Fighting Back). 6 tables and 26 references