Informing youth about the age of sexual initiation using means or percentages.

The common statistical indicator "mean age of first sex" can be misinterpreted by youth to indicate that most of their peers of the same age are sexually initiated, when this is not usually the case. This can jeopardize efforts to delay sexual initiation. University students were randomly assigned to one of two versions of an anonymous survey. They were asked to estimate how common sexual initiation was at a given age upon being presented with statements with different wordings, such as the "mean age of first sex" or "proportions of youth at different ages having had sex." Their interpretations were compared using logistic regression. Students who were assigned surveys using the indicator "mean age" of sexual initiation had higher odds of overestimating the extent of sexual initiation compared to those assigned surveys using percentages as the indicator, even after adjusting for student's sex and degree. We encourage the use of the "percentage" of youth, at different ages, who are sexually initiated as a more reliable indicator.