2-YEAR LONGITUDINAL-STUDY OF ALCOHOL EXPECTANCIES AND DRINKING AMONG NORWEGIAN ADOLESCENTS

Aims. The potential importance of alcohol outcome expectancies in the initiation and maintenance of drinking has been supported by studies s howing that these expectancies are present before drinking begins, and that they predict drinking both cross-sectionally and longitudinally. Although initiation of drinking behavior may be influenced by expecta ncy, subsequent drinking experience may modify expectations. We used s tructural modeling techniques to investigate the relative influence of expectancy and drinking in a three-wave longitudinal study of Norwegi an adolescents. Design. Survey incorporating self-administered questio nnaires. Setting. Twenty-two schools in Hordaland County on the west c oast of Norway. Participants. Nine hundred and twenty-four seventh-gra de students; 45.7% female. Measurements. Alcohol use (frequency, quant ity, drunkenness); the Norwegian version of the Alcohol Expectancy Que stionnaire for adolescents. Findings. Among students who were already drinkers upon entry into the study, expectations of positive social ef fects of alcohol predicted drinking longitudinally. Among those who be gan drinking during the study, these social expectancies predicted dri nking initiation, but drinking also influenced subsequent expectancy i n the early stages of drinking. Conclusions. These results support a r eciprocal relationship of drinking to positive expectancy, highlightin g the importance both of expectancies on influencing drinking, and of early drinking experiences on the development of positive expectancies .