"A bit more truthful": the validity of adolescent sexual behaviour data collected in rural northern Tanzania using five methods.

OBJECTIVE To assess the validity of sexual behaviour data collected from African adolescents using five methods. METHODS 9280 Tanzanian adolescents participated in a biological marker and face to face questionnaire survey and 6079 in an assisted self-completion questionnaire survey; 74 participated in in-depth interviews and 56 person weeks of participant observation were conducted. RESULTS 38% of males and 59% of females reporting sexual activity did so in only one of the two 1998 questionnaires. Only 58% of males and 29% of females with biological markers consistently reported sexual activity in both questionnaires. Nine of 11 (82%) in-depth interview respondents who had had biological markers provided an invalid series of responses about sex in the survey and in-depth interview series. Only one of six female in-depth interview respondents with an STI reported sex in any of the four surveys, but five reported it in the in-depth interviews. CONCLUSION In this low prevalence population, biological markers on their own revealed that a few adolescents had had sex, but in combination with in-depth interviews they may be useful in identifying risk factors for STIs. Self-reported sexual behaviour data were fraught with inconsistencies. In-depth interviews seem to be more effective than assisted self-completion questionnaires and face to face questionnaires in promoting honest responses among females with STIs. Participant observation was the most useful method for understanding the nature, complexity, and extent of sexual behaviour.