Patterns of alcohol consumption in Argentinean adolescents: Relationship with behavioral and personality variables

REYNA CE; SÁNCHEZ A; LELLO IVACEVICH MG

Montreal

Congreso; Society for Research in Child Development Biennial Meeting; 2011

The characterization of alcohol consumption is an area of considerable controversy. One of the most appropriate ways to identify typologies of substance use is latent class analysis, which can detect several problems of consumption rather than focusing exclusively on disease patterns (Reboussin, Song, Shresthab, Lohman & Wolfson, 2006, Smith & Shevlin, 2008). The identification of typologies not only has value in itself but especially when it aims to achieve a comprehensive understanding of alcohol consumption, namely in relation to other variables, for example: situational, environmental, cognitive, emotional, and personality. Several studies have shown uni and bidirectional relationships between drug use and aggressive behavior, though relations are not so clear when intervene factors such as the sex and age, let alone when personality variables are included, such as sensation seeking. In fact, a recent meta-analytic study notes moderate relationships between alcohol use and sensation seeking (Hittner & Swickert, 2006). Based on this, and because of the need for better understanding of consumer behavior of Argentinean adolescent, this study aimed to evaluate the relationship between the pattern of alcohol consumption and aggressive behavior and sensation seeking. The sample included 309 adolescents aged 12 to 19 years (M = 5.15, SD = 1.96), 35.6% male, from Cordoba city (Argentina). Alcohol consumption was assessed with questions about frequency, quantity, drunkenness and intent of consumption. We applied the Aggression Questionnaire of Buss and Perry adapted by Santisteban and Alvarado (2009), and the Sensation Seeking Scale of Zuckerman adapted by Smith (2008). Analysis of consumer behavior indicate that 55.1% of adolescents did not consume alcohol during the past 30 days, while the rest consumed at least 1 or 2 times per month. Based on current drinkers (44.9%), and from the indicators investigated, we developed a latent class analysis to identify consumer typologies. The model fit was evaluated primarily based on statistical BIC, AIC and AIC3. We identified the following classes: "binge" = 46.19% of the sample, "social" = 28.16% and "heavy" = 25.65%. Then, we performed a multinomial logistic regression analysis. The model shows an adequate fit (chi2 = 197.477, p