Affiliations: [a] Mathematica Policy Research, Inc., P.O. Box 2393,
Princeton, NJ 08543-2393, USA | [b] Agency for Health Care Policy and Research, Center
for Intramural Research, Division of Statistics and Research Methodology, 2101
E. Jefferson St., Suite 500, Rockville, MD 20852, USA
Note: [1] The views expressed in this paper are those of the authors and no official
endorsement by the Department of Health and Human Services or the Agency for
Health Care Policy and Research is intended or should be inferred. The
authors would like to thank Jane Faulman of Social and Scientific Systems,
Inc. for her programming support.
Abstract: Family-level data collected in longitudinal, as opposed to cross-sectional,
surveys present special analytical problems due to the dynamic nature of family
units over time. Family structure can change as a result of birth/adoption,
death, marriage, divorce, migration, and transient situations such as attending
college away from home. Different strategies for defining changing families for
analytical purposes have been proposed, including cross-sectional, longitudinal,
and dynamic approaches. In this paper, the authors will describe the extent and type of change in family
structure and composition in the Household Survey component of the 1987 National
Medical Expenditure Survey (NMES) over the course of the four data collection
rounds in the survey year and the way these changes are incorporated into
analytical plans. Several strategies for defining families which can change
across time will be employed and evaluated. A few annualized outcome measures of
interest at the family level will be explored using these various
definitions.
