|
Activity Number:
|
412
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 12, 2004 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #300979 |
|
Title:
|
Survey Errors and Survey Costs: Experience from Surveys of Arrestees
|
|
Author(s):
|
Yonghe M. Yang*+
|
|
Companies:
|
NORC, University of Chicago
|
|
Address:
|
1350 Connecticut Ave., NW, Washington, DC, 20036,
|
|
Keywords:
|
design effects ; survey errors ; survey costs ; cost models
|
|
Abstract:
|
The Arrestee Drug Abuse Monitoring program (ADAM) collects substance abuse data from samples of newly arrested criminals in 35 counties across the United States. The general survey literature contains little discussion of design and sampling issues involving arrestees. The purpose of this paper is to describe the sample designs adopted by ADAM. In particular, it discusses ADAM design options and decisions as they are related to survey errors and survey costs. Standard survey methods designed for the general population are not feasible for studying the highly clustered, highly skewed, and highly dynamic arrestee population. Common designs (e.g., sampling in time and space) developed for surveys of various mobile populations are not readily applicable either due to operational and budgetary constraints. Survey errors are measured by design effects arising from unequal weighting and clustering, and survey costs are estimated through explicit cost models. Nonsampling errors and administrative measures of survey quality are also considered.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2004 program |
