In many surveys, both national and local, people are required to answer questions about their past experiences. The data from these surveys are used in the formulation of legislation and other public policy purposes, and so they must be as accurate as possible. One goal of the present research is to explore some of the retrieval mechanisms that people use to answer questions about health-related experiences, and to design new ways of improving the accuracy of their responses. In all experiments, subjects will be members of a major health cooperative who will answer questions during a routine consumer opinion survey. When permission is granted, their records will be checked to verify the accuracy and completeness of their responses. Some specific questions addressed in this project are: (1) Do people who claim to have a good, rather than poor, memory for health-related experiences actually have better memories? (2) Does the order in which people retrieve health-related experiences influence the accuracy and completeness of their reports? (3) How good are people at dating their health-related experiences, and can techniques be developted to improve dating? (4) Does a person's current state of health influence recall? (5) How does proxy recall (recall about spouse, child or other person) compare to self recall? Data obtained will bear on current theories of information processing, and will also have the practical benefit of leading to recommendations for improved methods of gathering health-related information on surveys.
