|
Activity Number:
|
122
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2003 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301495 |
|
Title:
|
Methods for Combining Probability and Convenience Samples: Estimation, Inference, and Applications
|
|
Author(s):
|
Douglas Rivers*+ and Vicki J. Huggins and Daniel Slotwiner
|
|
Companies:
|
Stanford University and Knowledge Networks, Inc. and Knowledge Networks, Inc.
|
|
Address:
|
Dept. of Political Science, Stanford, CA, 94305,
|
|
Keywords:
|
|
|
Abstract:
|
Studies where the population of interest comprises a small percentage of the sampling frame are difficult to conduct, due to the cost and time necessary to screen participants for eligibility. In commercial applications, where cost considerations often predominate, convenience samples are frequently employed. Because little is known about the sample selection mechanism, the sampling distribution of sample statistics is unknown. Nonetheless, it is common practice to compute standard errors under the (incorrect) assumption of random sampling. This paper describes an alternative methodology that combines probability and convenience samples. Subpopulation characteristics are estimated based upon a small probability sample. This sample is supplemented with an inexpensive convenience sample with characteristics that match the estimated subpopulation characteristics. The methodology incorporates consistent tests for bias in the supplementary convenience sample, optimal weighting for combining the subsamples, and estimation of standard errors for the combined samples. Five case studies are presented to illustrate the methodology and assess its effectiveness.
|
- 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 2003 program |
