This paper will use the Survey of Income and Program Participation (SIPP) to show some illustrative differentials within the older population for the various age, sex, and race groups. We argue here that overall averages used to describe the older population are misleading and that data producers should show age, sex, and race differentials in the data to the extent possible. Because of sample sizes in surveys, such detail has not usually been feasible and so we suggest several statistical approaches to the problem.