Background: The purpose of this study is to compare the effectiveness of a national comprehensive teen pregnancy prevention (TPP) intervention to a national abstinence-only TPP intervention on middle school students' knowledge, attitudes, and behaviors related to teen sexual behaviors in a state with high teen birth rates. Methods: Pre- and post-intervention data were collected annually (2005-2010) from seventh-grade students to evaluate school-based TPP programs that implemented a comprehensive (N?=?3244) or abstinence-only (N?=?3172) intervention. Chi-square and t tests, logistic regressions, and hierarchical multiple regressions examined relationships between sexuality-related behavioral intentions, knowledge, and attitudes. Results: Students in both interventions reported significant (p?<?0.05) improvements post-intervention. Youth in the comprehensive TPP intervention were more likely (p?<?0.05) to have significantly improved their attitudes (odds ratios [ORs]?=?1.35, 1.83, 1.23) and behavior regarding abstinence decisions in the past 3 months (OR?=?1.39). The interventions' improvements in attitudes were more explanatory for behavioral intentions for students in the abstinence-only intervention than for students in the comprehensive TPP intervention. Conclusions: The mixed results suggest the comprehensive TPP intervention was only slightly more effective than the abstinence intervention, but that changing student attitudes and perceptions may be a key component of more effective TPP interventions.