RESUMO
OBJECTIVE: Our study was aimed at comparing health behavior data collected from a Web-based self-administered questionnaire (Web SAQ) versus a paper-and-pencil self-administered questionnaire and assessing the feasibility of the application. MATERIALS AND METHODS: One hundred and ninety (n = 190) pupils (ages 14-16 years) of senior high schools anonymously completed a questionnaire, with demographics and queries about lifestyle, alcohol, and tobacco use. For each class, the adolescents were randomly assigned to complete either the paper version of the questionnaire or the equivalent Web-based one, which used a customized platform developed for the purposes of this survey. RESULTS: Females who filled out the Web SAQ required significantly less time and completed a significantly higher percentage of its items. Although the majority of questions on tobacco and alcohol did not differ significantly across the two administration modes, there were gender-related differences in some sensitive information. Male adolescents on the Web SAQ accounted higher per hour drink consumption (r = 0.27, p = 0.015) and more numerous episodes of inebriety (r = 0.26, p = 0.010), whereas females seemed to state a younger age of alcohol onset (r = 0.33, p = 0.002). Females were more likely to report being monthly smokers on the Web SAQ (odds ratio = 0.37). Adolescents felt significantly less observed and females referred being more independent while compiling the Web SAQ. CONCLUSIONS: The findings of the study suggest that differences in reporting of some behavior of adolescents when using a Web SAQ do exist, despite the small-to-medium effect sizes. Exploiting the Web requires further investigation for extensive comprehension of the reasons for such differences.