Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)



Committee Chair(s)

Scott Bates


Scott Bates


Jamison Fargo


Julie Gast


D. Kim Openshaw


Karl White


Empirical investigations have identified hundreds of factors that predict whether youth engage in sexual activity (YSA). To promote optimal health and the avoidance of unhealthy or problematic outcomes that can result from YSA, sex education programs have been extensively developed and evaluated. Many evaluations have identified the effect of the program on immediate outcomes such as attitudes and intentions, others have examined subsequent behavioral and health outcomes, and some have done both. The purpose of this study was to extend the evaluation literature by testing a mediated effects model. A sex education program was found to have significant immediate effects on several attitudinal factors that have been shown to predict YSA, and was shown to significantly reduce the incidence of sexual activity approximately one year after the program (OR = 0.534, p = .004). A mediating effects test showed that youth’s stated intentions to engage in sexual activity was a significant mediated effect (B = -0.182, Lower CI = -0.291, Upper CI = -0.073), suggesting that the program effects on sexual activity occurred through the immediate effect on intentions, which in turn was likely affected by program content, which changed other attitudinal factors such as values, efficacy, and knowledge. Using immediate changes on these mediating factors to predict the likelihood of YSA showed that accurate prediction was possible, with an overall prediction accuracy rate of 74%. It was easier to predict who was not going to engage in YSA (94% accuracy) than who would (35% accuracy). Further predictive analyses showed that a score of 4.12 (on a scale of 1 to 5) on agreement with the items comprising the mediating factors’ scales was a threshold point, with the likelihood of engaging in YSA rising sharply as a function of this score until that point, and score increases above that point resulting in minimal changes in the probability of YSA. The results of this study demonstrate that it is possible to reduce YSA, that intent to engage in YSA was a primary mediator, and that accurate prediction of eventual behavioral results is possible, based on analysis of immediate results.




This work made publicly available electronically on September 2, 2011.

Included in

Psychology Commons