Measuring alcohol consumption: a comparison of the retrospective diary and the quantity-frequency methods in a college drinking survey.

As part of a larger study of college drinking, the quantity-frequency (Q-F) measure of alcohol consumption was compared to the retrospective diary (RD) method with 494 drinkers out of 606 respondents (75.8%) of 800 randomly selected undergraduates. Respondents were representative by sex and year in school of virtually all full-time undergraduates at Rutgers, the State University of New Jersey. The study tested the hypothesis that the RD would yield higher estimates of average alcohol consumption and of "heavy drinking" as defined by NIAAA typology (average two or more drinks per day) and of "peak drinking," a Q-F standard consisting of five or more drinks at least once per week. The study also compared the two methods as correlates of self-reported problems related to drinking. Results supported the major hypotheses that the RD would detect more overall consumption, heavy drinking and peak drinking than the Q-F. Results also demonstrated good consistency between the Q-F and RD as well as comparable correlations with the alcohol problem score.