Repeated Measures Estimation of Measurement Bias for Self-Reported Drug Use With Applications to the National Household Survey on Drug Abuse (From The Validity of Self-Reported Drug Use: Improving the Accuracy of Survey Estimates, P 439-476, 1997, Lana Harrison and Arthur Hughes, eds. - See NCJ 167

• NCJ Number: 167358
• Author(s): P P Biemer; M Witt
• Date Published: 1997
• Length: 38 pages
• Sale Source: National Institute on Drug Abuse; NCJRS Photocopy Services
• Publication Type: Issue Overview
• Language: English
• Country: United States of America
• Note: DCC

Annotation

This article reviews several methods for direct validity estimation that rely solely on repeated measures data to assess response bias.

Abstract

The article presents the results of an evaluation of a statistical method for estimating false positive and false negative reports from repeated measurement studies. The method was developed by Hui and Walter for evaluating medical diagnostic testing procedures. The article also presents a general model for studying misclassification in self-reported drug use and extends the model to the case where two measurements of the same characteristic are available for the sample of respondents. The article shows how the measures of reliability, measurement bias, estimator bias, mean squared error, false negative, and false positive probability can be defined in the context of the general model and how they may be estimated under the appropriate study designs. The article also demonstrates the use of Hui and Walter's method for evaluating the error in self-reported drug use. The article suggests that Hui and Walter's method should be considered for studies of drug use reporting error that use a biological test (hair, urine, or nail) to evaluate the error in the self-report. Tables, references