Validation of a measure of excessive drinking: frequency per year that BAL exceeds 0.08%.

Although self-report data are generally evaluated as reliable, validity estimates for self-reported drinking behavior are not as favorable. A new method of measurement addressing problems currently associated with the questionable validity of self-report drinking information is introduced. Study 1 tests the correspondence between the new measure (occasions per year blood alcohol level exceeded 0.08%), traditional self-report measures, and laboratory alcohol consumption. The novel measure correlated most strongly with volume of alcohol consumed in the laboratory compared to the traditional measures. In Study 2 the novel measure was favored over a traditional measure when discriminating between the presence and absence of problem-drinking symptoms.