Health Insurance Dynamics (HIDyn)
Data and Documentation
Introduction to the 1996 Panel of the Survey of Income and Program Participation
The 1996 Panel of the Survey of Income and Program Participation (SIPP) is one of a continuous series of national panels conducted by the U.S. Bureau of the Census to collect detailed data on income, employment, and government program participation. Each SIPP panel is a true longitudinal survey that tracks individuals over time. The data is released with one record for each month for each participant. The 1996 panel is the only panel containing 12 waves of interviewing over 4 years; other panels usually cover 2 1/2 to 3 years. The first SIPP panel was fielded in 1984; the 2001 panel is currently ongoing.
The Census Bureau uses a two-stage sample design to select geographic primary sampling units (PSUs) and then address units within sample PSUs to identify SIPP sample members. The resultant clustering necessitates adjustment for sample design effect in variance estimates. Sample strata indicators provided by SIPP for this purpose are included in HIDyn. The 1996 panel sampled approximately 40,000 households, in which all household members age 15 and older are interviewed if possible. Proxy responses are collected when self-response is not possible and for children under the age of 15 years. Importantly, over-sampling of low-income and minority populations ensures adequate sample size for their analysis and requires the use of SIPP weights when analyzing the entire sample. Weights adjust subpopulation sizes to proportions observed in the population and adjust for sample non-response. The Census Bureau handles item non-response by imputation. Allocation flags for imputed items are included in HIDyn.
The SIPP follows all original sample members, including those who move within the United States, if they are not institutionalized or living in military barracks, interviewing each respondent every 4 months. When original sample members move into households of individuals not previously in the survey or when such individuals move into the households of original sample members, the new individuals become part of the SIPP sample. A core set of questionnaire items for measuring the respondent's economic situation repeats at each 4-month interview wave, using a 4-month recall period to collect data for the preceding months. Modules of questions not addressed in the core interviews are asked at various interviewing waves. Separate topical modules cover personal history, child care, wealth, program eligibility, child support, disability, school enrollment, taxes, and annual income. Computer-assisted interviewing is used in conducting face-to-face personal interviews (CAPI) and decentralized telephone interviews (CATI).
Further details about SIPP's design and use, including the latest edition of the SIPP User's Guide are found at the U.S. Bureau of the Census' 1996 Survey of Income and Program Participation web site.
Last modified: 06/21/06 | Contact Webmaster
