The Economic Fortunes of Women and Children: Lessons from the Panel Study of Income Dynamics

Abstract

Most of our knowledge about the economic fortunes of women and children comes from case studies, special-purpose surveys such as National Longitudinal Surveys, or cross-sectional data. Each source has certain limitations. Case studies often provide richly detailed pictures of a few individuals who may not, however, be representative of larger groups. Analyses of special-purpose longitudinal surveys may yield generalizations about only limited segments of the population. Cross-sectional data provide snapshot pictures of an entire population but tell us little about the dynamic processes that affect their lives. Because most social science theories of human behavior are dynamic rather than static, they must be tested through repeated observations of the same people. An extra year of data on one set of individuals is considerably more valuable than new data on a different cross section. This is particularly important when the process under analysis takes an extended period of time. For instance, investigations of the causes, extent, and consequences of long-run welfare dependency must be conducted over an extended period and even across generations if competing theories are to be tested adequately. This requirement also holds for studies of changes in family composition (e.g., marriage, childbirth, marital breakups), which should be based on comparisons of a family's situation before and after change. Equally important, any study of the economic well-being of women