An Experiment to Reduce Measurement Error in the SIPP: Preliminary Results
An Experiment to Reduce Measurement Error in the SIPP: Preliminary Results
Working Paper Number: SEHSD-WP1994-07 or SIPP-WP-197
A record check study conducted on the 1984 Survey of Income and Program Participation (SIPP) in 4 states, showed underreporting rates for 3 income programs at around 25% and even higher for the unemployment insurance program (see Figure 1). These errors not only distort estimates of means and percentages but also, as Bollinger and David (1993, 1994) have shown, distort estimates of relationships in important policy models that rely on SIPP data.
This led us to design new interviewing procedures with the goal of reducing SIPP underreporting error by at least 25% for these major transfer income programs.
Others in Series
Working Paper
A Comparative Analysis of the Labor Force Activities
An analysis of the labor force activities of men and women from the black, American Indian, Asian, Mexican, other Hispanic and non-Hispanic white populations.
Working Paper
Changing Social Security Benefits to Reflect Child Care Years
This paper estimates the effects of child care proposals for women categoried by economic and demographic characteristics.
Working Paper
Longitudinal Imputation of SIPP Food Stamp Benefits
This research examines the current and three alternative longitudinal imputation methods as applied to the SIPP item of food stamp income.
Related Information
Page Last Revised - October 8, 2021