Affiliations: [a] Mathematica Policy Research, Inc., 600 Maryland Ave.
S.W., Suite 550, Washington, DC 20024-2512, USA
Note: [1] Prepared for the Conference on the Future of SIPP sponsored by National
Academy of Sciences, Committee on National Statistics, Washington, D.C.
April 11–12, 1991. The author greatfully acknowledges comments received from
Harold Beebout, Constance Citro and Denton Vaughan.
Abstract: The government sponsors over 20 voluntary cash and in-kind programs to assist
needy persons. To measure program effectiveness, characteristics of the target
population, and the impact of program reforms, the government requires a survey
that supports estimation of the size and composition of the eligible population.
This paper critiques the utility of the Survey of Income and Program
Participation (SIPP) for measuring eligibility under 5 of these need-tested
programs. SIPP offers unique advantages over other surveys: monthly accounting
period, periodic measures of assets and expenses, and repeated measures of
household composition, program unit composition, and income from over 50
different sources. SIPP has disadvantages as well: small sample, omission of
some determinants of program eligibility, lack of repetition of key data
elements, and data that have been subject to errors and delays. The small sample
size is the most crucial barrier to wide-spread use of SIPP for program
analysis.
