Survey of Program Dynamics (SPD), 2002: Cross-Sectional File (ICPSR 3954)

Published: Dec 15, 2005

Principal Investigator(s):
United States Department of Commerce. Bureau of the Census


Version V1

SPD 2002

The 2002 SPD Cross-Sectional file provides socioeconomic data for each household member for calendar year 2001. Demographic data focuses on age, sex, race, ethnic origin, marital status, household relationship, education attainment, and veteran status. Economic data focuses on work experience, including comprehensive work experience, as well as the employment status, occupation, industry, weeks worked, hours per week worked, total income, and income components for people aged 15 and over. Income data covers income from jobs, net income from businesses, farm, or rent, and income from pensions, dividends, interest, and social security payments. Data are also included on noncash income sources, such as food stamps, school lunch programs, employer-provided group health insurance plans, employer-provided pension plans, personal health insurance, Medicaid, Medicare, CHAMPUS/Tricare or military health care, and energy assistance.

United States Department of Commerce. Bureau of the Census. Survey of Program Dynamics (SPD), 2002:  Cross-Sectional File. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2005-12-15.

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote


2002-05 -- 2002-07

(1) The 2002 SPD Cross-Sectional file is minimally edited and is intended for analyses of effects of welfare reform on individuals, families and households. (2) The file is a rectangular, person-level file. Household- and family-level variables are included on the record for every person in the household.

Due to budget constraints, the SPD 2002 Survey did not visit all of the households eligible for interview (including approximately 30,588 interviewed, and Type A, Type B, and Type D noninterviewed households in the SPD 2001 sample and their spawned households). The budget allowed for SPD to visit only 20,000 households. Thus, a sample cut was carried out similar to the 1998 sample cut. Roughly 19,100 households were sampled in this operation, in order to account for an expected net five percent growth rate due to household spawning. Interviewed and Type A, Type B, and Type D noninterviewed households were eligible to be selected. The eligible households were classified into six strata and then randomly selected for the SPD 2002 sample such that each stratum took an approximately equal proportional cut to reduce the total sample size to approximately 19,100.

The universe consists of people residing in the United States, except people living in institutions, such as prisons, and nursing homes, and entire military households, in March 2002.


survey data

computer-assisted personal interview (CAPI)



The weights on this file are cross-sectional, therefore, they are only valid for estimates of the characteristics of the cohorts of people in 2002 who were represented by both the original and nonoriginal sample people in the SPD 2002. An original sample person is a sample person who was a self or proxy respondent in Wave 1 of the SIPP 1992 or 1993 panel. A nonoriginal sample person is a sample person who became part of the sample after the SIPP Wave 1 and beyond (including the SPD samples). The SIPP 1992 and 1993 Panel samples, from which the SPD sample originated, were designed to produce only national estimates of the characteristics of interest to the user. Although this 2002 cross-sectional file includes state identifiers and is weighted to current state (as well as national) controls, deriving subnational estimates from this nine-year-old sample that has experienced severe attrition (household and person nonresponse) and other sample loss (through death, being institutionalized, etc.), as well as sample cuts dictated by budgetary constraints, results in inadequate representativeness and poor coverage in some of the smaller geographic areas. Even for the national estimates, inadequate representativeness and poor coverage occur in Black children aged 11 and below and Black adults aged 20 to 45. In addition to noninterview adjustment, we attempted to further correct for these deficiencies by bringing back noninterviews from the 1997 Bridge sample and from a subset of the SIPP 1992 and 1993 Panels. These sample cases are included in this cross-sectional file, and will be included in the third (and final) longitudinal file, as well.


  • Data in this collection are available only to users at ICPSR member institutions.

ICPSR logo

This study is provided by ICPSR. ICPSR provides leadership and training in data access, curation, and methods of analysis for a diverse and expanding social science research community.