The Survey of Income and Program Participation (SIPP) Methods Panel Improving Income Measurement

The Census Bureau recently established the Survey of Income and Program Participation (SIPP) 2000 methods panel project to evaluate and redesign the core instrument for SIPP, a recurring, nationally representative, longitudinal survey of people and their socio-economic characteristics. The objectives of the project are to improve response rates in SIPP, to reduce income under reporting, and to improve data quality. It is a research project consisting of analysis of extant data as well as experimental research. The data analysis component includes examining patterns of nonresponse, examining reporting patterns across waves of interviewing, and analyzing patterns of income receipt. The program of experimental research consists of three phases, designed to allow for two iterations of testing and refining the Wave 1 and Wave 2+ core instruments. Each phase will culminate in a split-sample field test of about 2,000 interviewed Wave 1 households—1000 randomly assigned to each of the control treatment (receiving the standard SIPP instrument) and experimental treatment (receiving a modified instrument). We will assess our success in meeting our objectives through analysis of data obtained in the field experiments and debriefings of respondents and interviewers.