This table presents analyses of sample representativeness for the Health and Retirement Study (HRS), contrasting both weighted and unweighted estimates for common variables across the American Community Survey (ACS), Current Population Survey (CPS), and the National Health Interview Survey (NHIS). Our analysis used 2020 data from all sources and was restricted to non-institutionalized individuals aged 55 and older, aligning with the youngest age cohort in HRS for that year.
Weighted estimates were calculated using the final survey weights provided in each dataset. We computed 95% confidence intervals for each descriptive parameter and examined the overlap of these intervals across the data sources.
| Variable Name | Data Source | Variable Response Category | Weighted | Unweighted | ||||
|---|---|---|---|---|---|---|---|---|
| Mean/Prop | 95 % CI | Mean/Prop | 95 % CI | Lower Bound | Upper Bound | Lower Bound | Upper Bound | |
Notes
- In CPS, 80-84 was coded as 80, 85+ was coded as 85; In NHIS, 85+ was coded as 85.
- CPS and NHIS did not provide birth year, so the age cohort used were calculated using the age at the time of the interview.
Suggested Citation
- Zhao C, Schroeder H, West BT. HRS Sample Representativeness Estimates. Ann Arbor, MI: Survey Research Center, Institute for Social Research, University of Michigan; 2025