Measuring Alcohol Consumption in the United States

Abstract

This chapter lays out the methods of measuring alcohol consumption that have been used in survey studies of the general population of the United States and discusses their development and rationales. As was already clear by 1970 (Room, 1970b), there are two major strategies for asking respondents about their current pattern of drinking: (1) asking them to list all recent drinking occasions; and (2) asking them to summarize their current patterns. In general terms, British and Scandinavian researchers have followed the first strategy, while North American researchers have followed the second (see Room, 1977, and Auth and Warheit, 1982/83, on U.S. traditions). Recently, some European researchers have challenged U.S. traditions of measuring drinking patterns, arguing in favor of the “recent occasions” approach (Duffy, 1982, 1984; Alanko, 1984). On the other hand, a recent empirical analysis of Finnish data concluded that “for most descriptive purposes” on a population level, “the choice of measurement procedure is irrelevant,” and furthermore that an approach in the U.S. tradition may least underestimate consumption among the heaviest drinkers (Simpura, 1987). These discussions and analyses, along with recent discussions by U.S. researchers (e.g., Greenfield, 1986; Knupfer, 1987a, 1987b), do remind us of “the extremely complicated structure of the concept of alcohol consumption” (Alanko, 1984, p. 209) and of the need for new developmental work taking account of the different national traditions. As Alanko concludes, our “awareness of the problems” in existing approaches “should stimulate further research leading to improved methodology” (p. 224).