Variables Related to Inconsistent Answers in a Survey About Sexual Intercourses Among Students in Santa Marta, Colombia

Abstract

Background: Self-reported questionnairesare the more common and reliable method toinvestigate health-related behaviors among adolescents.Often, a significant number of participantsreport inconsistencies in several relateditems, especially sensitive topics, and they mustbe omitted. Up to date, it has not extensivelycompared the demographic characteristics ofstudents that answer consistent and not consistentlya paper-pencil survey.Objective: To compare some characteristicsrelated to inconsistent answers about sexualbehavior among students of secondary school atSanta Marta, Colombia.Method: A probabilistic cluster simple of studentscompleted an anonymous survey aboutsexual relationship. Logistic regression was usedfor adjusting associated variables to answer inconsistentlyquestions about sexual behavior.Results: A total of 3813 students completedthe survey. A group of 3575 participants (93.8%)reported consistent information about sexualbehavior; and 238 (6.2%) did not. Studentsthat completed inconsistently the survey aboutsexual behavior were more frequently attendingto private school (OR=3.5; 95%CI 2.6-4.8) andboys (OR=2.1; 95%CI 1.6-2.8), after adjustingfor socioeconomic level.Conclusions: Approximately one out oftwenty students completed inconsistently inquiresabout sexual behavior. Inconsistent answersabout sexual behavior were higher in privateschool and male students. More investigationis needed.

Keywords

• Sexual behavior
• questionnaires
• students
• adolescents
• cross-sectional study