ABSTRACT
A great deal has been learned in recent years about age-related cognitive performance and underlying processes, much of it through laboratory and clinical research. More recently, measures of cognitive performance have been included in several large-scale sample surveys. There appear to be at least two major reasons for the inclusion of cognitive performance measures in surveys. First, descriptive information about cognitive functioning and its natural course in the older population as well as the predictors and consequences of cognitive functioning are still needed because experimental and clinical investigations are not well suited to providing representative population estimates. For example, findings are inconsistent about whether crystallized intelligence or knowledge declines with age, an inconsistency that might possibly reflect the use of nonprobability samples in experimental cognition research (Salthouse, 1997). Or, information is sparse on how cognitive impairment and decline impacts on the daily functioning of older people, on their need for formal health care and informal assistance, and on their economic resources. These are all areas where survey research can make unique contributions.
