The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #66972

James Wiley, PhD¹, Denise Zabkiewicz, MPH², and Laurie Jacobs, MA². (1) San Francisco State University, 1600 Hollaway Avenue, San Francisco, CA 94132-4025, 415-338-6716, jwiley3@sfsu.edu, (2) University of California, Berkeley, Department of Epidemiology, Public Health Institute, Alcohol Research Group, 2000 Hearst Ave., Suite 300, Berkeley, CA 94709

Self-report and administrative records are two common sources of data on service use in welfare populations. Yet, they represent two versions of welfare reality: the reality created by the welfare bureaucracy and the reality experienced by the client. Findings from previous research suggest that there is no “gold standard” for data on service encounters. The purpose of this study is to examine the degree of correspondence between these two sources of data in order to assess their properties for the measurement of service delivery. Using one-year follow-up data collected in 2002 from a large California county sample of aid recipients (n=718), we match retrospective 12-month self-report histories of aid and services receipt, obtained utilizing timeline follow-back calendar techniques, with administrative records from the study county’s Department of Social Services. Using correlation analysis and Kappa measures of agreement, we examine the correspondence between month-to-month services as well as aggregate aspects of welfare service trajectories (such as the number of months of aid or treatment received). We compare measures of agreement by substance abuse status to assess the role of substance abuse om accuracy of recall. In addition, we examine the role of “telescoping”, the notion that events nearer in time are more clearly remembered than later events. Conclusions drawn from this analysis have implications not only for interpretation of empirical results but, for the methods and data utilized in analyses that examine factors related to service utilization more generally.

Learning Objectives:

• Examine the extent of correspondence between self-report and administrative records data in order to assess their properties for the measurement of welfare service delivery.

Keywords: Methodology, Welfare Reform

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.