SIPP Home > Introduction to SIPP > Comparison of SIPP with Other Surveys >
Because there is some overlap in the content of SIPP and certain other surveys, the question arises: When should an analyst use SIPP instead of the other surveys? A brief look at selected surveys might provide some guidance (Table 1-1 compares some key points as well).
The CPS, sponsored jointly by the Census Bureau and the Bureau of Labor Statistics (BLS), is primarily a labor force survey. It is used to compute the federal government.s official monthly unemployment statistics, along with other estimates of labor force characteristics. In addition to its core content, a different supplement is fielded each month. One of these, the March Annual Demographic Supplement, is currently the official source of estimates of income and poverty in the United States. Compared with SIPP, however, the CPS has gaps in the area of income measurement. A yearlong reference period means that CPS respondents are more likely than SIPP respondents to forget or misreport certain asset income or irregular income sources. The CPS does not collect data on assets and liabilities to the same extent as SIPP. The CPS is also less comprehensive in the area of program participation, sometimes missing partial-year data. The CPS reporting unit is the person, but the sample covers housing units; whoever happens to be living at the address at the time of the interview is in the sample. When residents of a CPS housing unit move, they are not followed; instead, the new residents become sample members. Housing units spend 4 months in the sample, 8 months out, and 4 months in again. The target sample size for the CPS is 50,000 housing units each month. Like SIPP, the CPS sample covers the U.S.-resident noninstitutionalized population, although, unlike SIPP, the CPS includes people living in military barracks.
| Feature | Survey of Income and Program Participation | CPS (March Income Supplement) | Panel Study of Income Dynamics |
| Sample size and design | 1996 Panel: 40,188 households; new panel periodically; each original- sample adult in panel for no. of months in survey; interviews every 4 months | 50,000 households; each household in sample for 8 months over 2-year period; rotation group design; monthly interviews (income supplement once per year) | 9,000 families; over- represents low-income families; continuing panel with annual interviews |
| Sample designed to be representative within states? | No | Yes | No |
| Income data | Data for about 70 cash and in-kind Sources at each 4- month wave, with monthly reporting for most Sources | Data for prior calendar year for about 35 cash and in-kind Sources | Data for prior calendar year for about 25 cash and in-kind Sources with specific months received |
| Tax data | Information to determine federal, state, and local income taxes; payroll taxes; property taxes | None | Information to determine federal, state, and local income taxes; payroll taxes; property taxes |
| Asset-holdings data | Detailed inventory of real and financial assets and liabilities once each year for panels from 1996 forward and at least once per panel in prior years; more frequent measures for assets relevant for assistance programs | None, except home ownership | Regularly, information about home value and mortgage debt; occasionally, information about saving behavior and wealth |
| Expenditure data | Information at least once each panel before 1996 and once a year 1996 and beyond on previous month.s out-of-pocket medical care costs, shelter costs (mortgage or rent and utilities), dependent care costs, and child support payments | None | Monthly rent or mortgage costs; annual utility costs; average weekly food costs; child support payments |
|
Main |
Introduction to SIPP |
SIPP Survey Content |
Technical Information |
Using & Linking Files |
SIPP Publications |
|
Access SIPP Data |
SIPP Users' Guide |
SIPP Tutorial |
User Notes/ListServe/News |
SIPP Help |
Page Last Modified: May 9, 2006
