Health Insurance Dynamics (HIDyn)
Data and Documentation
Longitudinal and Cross-Sectional Analyses of HIDyn
SIPP is designed for cross-sectional analysis by month or longitudinal analysis over the length of the panel. Because all of the data released by the Census Bureau for any person in any month are included in HIDyn, the reorganized database can also be analyzed cross-sectionally or longitudinally by making use of the weights and instructions for weighting and variance estimation provided by the Census Bureau with the public use files.
However, the great strength of HIDyn is for longitudinal analysis. In constructing the longitudinal weights, the Census Bureau restricted the population and the sample to people in the non-institutionalized population of the United States at the start of the survey. As a result, newborns, immigrants, and others who subsequently joined the target population were excluded for purposes of longitudinal analysis.
Although the longitudinal weight provided by the Census Bureau adjusts for differential non-response and attrition associated with many characteristics (including age, gender, ethnicity, race, household/family structure, employment, family income and assets, welfare receipt, education, and geographic location), the longitudinal weight does not specifically consider health insurance status. Consequently, for its analyses the HIDyn team made a final adjustment to the Census Bureau's longitudinal weight to match weighted counts of health insurance status by age and family income in the last month of the survey according to the monthly weight also provided by the Bureau.
The post-stratification of the revised weight was based on weighting cells defined by health insurance status (hierarchically assigned as Medicaid, Medicare, military/CHAMPUS, private, and uninsured), age (4-12, 13-18, 19-24, 25-34, 35-44, 45-64, 65+), and family income (above and below 200% of the poverty line) in Month 48. This adjustment was made after comparisons involving the public use weights showed that the cross-sectional estimate of the number of uninsured in the last month was higher for the population 4-64 years old according to the monthly weight (34.2 million, 15.2% of the population) compared to the longitudinal weight (30.7 million, 13.7% of the population). The longitudinal weight was post-stratified to the monthly weight out of concern that lack of health insurance was statistically associated with other changes in life circumstances that caused members of the longitudinal cohort to move and be lost to follow up. The research team judged that the representation of movers is better in the monthly sample than the longitudinal sample, because people who move into SIPP households during the survey have positive monthly weights and are included in monthly estimates.
This link takes you to the data dictionary provided by the U.S. Bureau of the Census for person-month variables in the 1996 Longitudinal SIPP File. This data dictionary contains one entry corresponding to 48 variables in HIDyn. For example, TAGE in the SIPP file corresponds to TAGE1-TAGE48 (one variable for each survey month) in HIDyn. The HIDyn Data Description includes the original variable's position on the Census Bureau's longitudinal SIPP file to facilitate cross referencing.
Last modified: 06/21/06 | Contact Webmaster
