Use of Sampling Weights with IPUMS MEPS
The target population associated with a Medical Expenditure Panel Survey Household Component (MEPS-HC) survey year is the U.S. civilian, noninstitutionalized population for a given year. Annual MEPS-HC sample weights for the household component are provided in the yearly Full Year Consolidated Data Files. Weight variables are provided for generating estimates of totals, means, percentages, and rates for persons and families for this population using the MEPS-HC.
There is a single full year person-level weight (PERWEIGHT) assigned to each record for each key, inscope person who responded to MEPS for the full period of time that he or she was inscope during the survey year. A key person was either a member of a responding NHIS household at the time of interview, or joined a family associated with such a household after being out-of-scope at the time of the NHIS. These newly inscope persons include newborns as well as those returning from military service, an institution, or residence in a foreign country. A person is inscope whenever he or she is a member of the civilian, noninstitutionalized portion of the U.S. population.
The person-level weight PERWEIGHT was developed in several stages:
- Person-level weights were created for each of the two panels in a survey year separately. The weighting process for each panel includes adjustments for nonresponse over time, and a calibration to independent population totals. The calibration was initially accomplished separately for each panel by raking the corresponding sample weights to Current Population Survey (CPS) population estimates based on five variables: census region (Northeast, Midwest, South, West); MSA status (MSA, non-MSA); race/ethnicity (Hispanic; Black, non-Hispanic; Asian, non-Hispanic; and other); sex; and age.
- A survey year composite weight was then calculated by multiplying each individual panel weight by a factor which reflects the relative sample size of the individual panel compared to the sample size for the two panels combined. Using such factors to form composite weights serves to limit the variance of estimates obtained from pooling the two panels.
- The composite weight resulting from the above procedure was raked to the same set of CPS-based control totals.
- When the poverty status information (derived from MEPS-HC income variables) became available, another raking was undertaken using dimensions reflecting poverty status in addition to the previously mentioned five variables. Thus, the raking for the final weight reflects poverty status in addition to the five variables mentioned above.
Self-Administered Questionnaire Weights
Beginning in 2000, a Self-Administered Questionnaire (SAQ) was included in the MEPS. Each adult person (age 18 or older) completed this questionnaire. Thus, the target population for the SAQ is adults in the civilian, noninstitutionalized population at the time data were collected for rounds 2/4. The SAQ weights are constructed by first adjusting the person weights for questionnaire non-response. Variables used in the nonresponse adjustment process are region, MSA status, family size, marital status, level of education, health status, health insurance status, age, sex, and race/ethnicity. Then, before 2011, the weights are raked to Current Population Survey (CPS) estimates corresponding to December of the survey year (the same source of control figures used for the full year person weights). As mentioned above, for 2011 and later, the weights were raked to the 2010 census-based population estimates. The variables used to form control figures (region, MSA status, age, sex, and race/ethnicity) are the same variables that were used for the full year person weights. The only difference was that age categories were developed after excluding ages under 18, since only adults were eligible for the SAQ. The two raking efforts are used for the Consolidated file in order to maintain consistency with how the sample weights were computed in previous years. The final SAQ weight is then obtained by raking the preliminary weight to CPS estimates that were based on poverty status as well as the aforementioned variables. This weight is assigned the variable name SAQWEIGHT.
Diabetes Care Survey Weights
The Diabetes Care Survey (DCS) was also added to MEPS in 2000. A person level weight, DIABWEIGHT was developed for use with the data obtained from the DCS. This weight was assigned to each person with a SAQ weight who was also classified as having diabetes (age 18 and older).Note: This user note is based on the documentation accompanying the 2014 Medical Expenditure Panel Survey Full Year Consolidated File (H-171), created by staff at the Agency for Healthcare Research and Quality, Center for Financing, Access, and Cost Trends.