Are Key Informant Estimates of Health-Risk Behavior Really More Valid Than Self-Report Estimates?

Abstract

In their article in the August, 1984 issue of this journal, Deaux and Callaghan (1984) conclude that estimates of health-risk behavior obtained from key informants are "apparently more valid than those produced by a telephone survey" and hence that "the key informant approach is superior to the RDD telephone survey." This conclu sion goes far beyond the data presented. What has in fact been demonstrated by Deaux and Callaghan is that different estimates of health-risk behavior were obtained on the same population using the two different methods. The authors assume that the key informants' estimates are ipso facto more valid measures of health-risk behavior without providing convincing evidence to support this assumption.