Sample Design and Estimation in the National Social Life, Health, and Aging Project: Round 3 (2015-2016)

Area of Science:

• Gerontology and Demography
• Social Sciences
• Public Health

Background:

• The National Social Life, Health, and Aging Project (NSHAP) is a key longitudinal study on aging in the United States.
• Previous data collection rounds established a foundation for understanding health and social factors in older adults.
• The 2015-2016 data collection aimed to expand the study's scope and representativeness.

Purpose of the Study:

• To provide the methodological framework for the 2015-2016 NSHAP data collection, including sample design and weighting.
• To detail the reinterview protocol for the original cohort (C1) and the sample design for the new cohort (C2).
• To enable robust statistical analysis of NSHAP data for both cross-sectional and longitudinal research on aging.

Main Methods:

• Reinterviewed the original cohort (C1, born 1920-1947) and initiated data collection for a second cohort (C2, born 1948-1965).
• Employed a probability-based sample design for C2 using the NORC U.S. National Sampling Frame, including stratification and clustering.
• Constructed cross-sectional weights for individual and couple-level analyses, accounting for the complex survey design.

Main Results:

• Successfully conducted 2,409 interviews with C1 respondents, achieving a 91% conditional response rate and a 71% unconditional three-wave response rate.
• Completed 2,368 interviews with C2 respondents, yielding a 76% response rate.
• Combined C1 and C2 samples represent the U.S. population of home-dwelling adults aged 50-96 in 2015-2016.

Conclusions:

• The 2015-2016 NSHAP data collection successfully expanded the study's cohort and provides a valuable resource for aging research.
• The combined C1 and C2 data permit inference about the U.S. population of adults born 1920-1965.
• Availability of three waves of C1 data enables longitudinal analyses of health outcomes and determinants among older adults.