Tobacco use by black and white adolescents: the validity of self-reports

Previous studies concluded that Black adolescents use tobacco and other drugs less than White adolescents. The Black-White differences typically were attributed to variations in background and life-style. The objective of the research reported in this paper was to determine whether the presumed difference in tobacco use is due to Black-White differences in the validity of self-reports. We used biochemical measures to compare the validity of self-reports of tobacco use by 1823 Black and White adolescents and to assess the contribution of variation in validity to Black-White differences in reported tobacco use. The sensitivity of Blacks' reports was significantly less than the sensitivity of Whites' reports. The specificity of Whites' reports was significantly less than the specificity of Blacks' reports. Much of the Black-White differences in reports of cigarette smoking and tobacco use were due to Black-White differences in validity. Studies of Black-White differences should adjust for the invalidity of reports or acknowledge that much of the difference may be due to measurement error.