SIPP Sample Construction
In the terminology of the Survey of Income and Program Participation, a
Panel refers to a group of interviewees and the year in which that group was
first interviewed. A Wave refers to an interview within a Panel. Each Panel
has anywhere from 3 (1989 Panel) to 12 (1996 Panel) interviews. Further
details of the SIPP can be found on the Internet at
Combining data across Panels was not possible for sample years 1986, 1989, 1990, 1995, 1996, and 1997 due to the lack of availability of overlapping Panels for which disability information is available. The labor market information, due to the rotation of sets of questions in the SIPP, corresponds to June of the year referenced. The exception to this pattern is the 1997 sample labor market information, which came from Wave 5 of the 1996 panel, which corresponds to August of 1997. Of course, these waves were merged with the disability topical module for that panel.
The disability check in the topical module was the variable used to designate a work-limiting disability. There are two checks in topical modules for the 1986 —1993 Panels, but only one check in the 1996 Panel, resulting in the slightly smaller incidence of individuals with work-limiting disabilities in the 1996 and 1997 samples. The match rate across Waves within a Panel ranged from 81 to 89 percent success. Labor status refers to activity during the previous month (as opposed to the previous week, as in the CPS), since job information corresponds to activity over the month.
The primary usefulness of the SIPP derives from an ability to identify the nature of a disabled worker's disability. The categories identified are too numerous for all of them to be included in the analysis, so they are grouped into broad headings based on the classifications used by the Social Security Administration. Table B.2 shows how specific disabilities are classified.