Skip Navigation
Administration for Children and Families  
ACF
ACF Home   |   Services   |   Working with ACF   |   Policy/Planning   |   About ACF   |   ACF News   |   HHS Home

  Questions?  |  Privacy  |  Site Index  |  Contact Us  |  Download Reader™  |  Print      

Office of Planning, Research & Evaluation (OPRE) skip to primary page content
Advanced
Search

     

Overview

The Earnings of Welfare Leavers: Substantive and Methodological Explorations demonstration will examine two methodological issues in the context of the Leavers Studies conducted in the CalWORKS demonstration. One issue is to determine how well data from unemployment insurance (UI) records reflect total family earnings and income from all sources. The other issue is to determine whether outcomes for persons that do not respond to surveys can be accurately estimated from those that do respond.

This demonstration addresses two major research questions. The first question involves determining the extent to which earnings, as recorded in UI records, reflect the real earnings of a particular recipient and how well UI earnings reflect the total earnings and income of the family. The second question involves determining how representative survey respondents are, whether non-response bias can be fixed through proper weightings, and what data should be used to construct these weightings.

In order to determine the extent to which UI data reflect total family income, several databases for recipients in the CalWORKS demonstration will be linked. The CalWORKS demonstration involves a recipient interview that will probe for employment and earnings that might not be reflected in the UI records. In addition, records from the California Department of Social Services will be linked to determine other sources of family income such as TANF payments and Food Stamps. Finally, RAND will obtain SSA data to learn about earnings sectors not covered by UI records. In order to determine whether non-response bias can be eliminated, data from surveys will be linked to other databases that contain outcomes of interest. For outcomes that can be determined outside the survey, comparisons can be made with techniques that seek to extend survey results to the non-response group. The demonstration will test whether propensity score techniques can be used to accurately predict outcomes for non-respondents.

The project period is 9/30/00-12/31/04 and the point of contact is Leonard Sternbach. Ask a Question.