As discussed in Chapter II, participation was elective in two sites (ReCapturing the Vision and FUPTP), leading some youth not to participate despite being randomly chosen for the program group. The implication of this nonparticipation is that the estimates of program impacts for the subsample of participants will be larger than the estimates for the entire sample in these two sites. However, because there is a corresponding loss of statistical power when estimating impacts for the smaller, participant-only sample, the statistical significance associated with these participant-only impacts is roughly equal to those for the full sample. Thus, the benefit of examining impacts for the participants-only sample is merely in identifying any notable program-control group differences, regardless of significance, that might have been less evident for the full sample.
As highlighted by Table IV.6, the estimated impacts on a few of the behavioral measures are somewhat notable in size for the participant-only sample, although they are not statistically significant. (For complete results of the participant-only impact analysis, see Appendix Table A.9.) With regard to sexual abstinence, for example, participants in ReCapturing the Vision had rates 7 percentage points higher than their control group counterparts. In contrast, program group youth in FUPTP reported rates of sexual abstinence that were 4 percentage points lower than their control group counterparts and 3 points lower when measured as abstinent in the last 12 months.
|My Choice, My Future!||ReCapturing the Vision||FUPTP||Teens in
|Estimated Impacts for Full Program Group|
|Remained abstinent (always)||1||5||-3||-4|
|Abstinent last 12 months||1||5||-2||-2|
|Expectations of Future Behavior|
|Expect to abstain through high school||5||6||-4||-1|
|Expect to abstain until married||-4||7||10**||-3|
|Estimated Impacts for Participants Only|
|Remained abstinent (always)||1||7||-4||-4|
|Abstinent last 12 months||1||8||-3||-2|
|Expectations of Future Behavior|
|Expect to abstain through high school||5||9||-6||-1|
|Expect to abstain until married||-4||11||18**||-3|
|Sample Size Total||448||480||414||715|
|Program Group Total||286||275||274||374|
|Program Group Participants||286||180||157||374|
|Source: Wave 4 Survey of Teen Activities and Attitudes (Mathematica Policy Research, Inc., 2005), administered to youth 42 to 78 months after enrolling in the Title V, Section 510 Abstinence Education Program study sample.
Note: All estimates are based on weighted regression models. For details on the covariates, see Appendix Table A.1. For complete results for the subgroup see Appendix Table A.9. Study youth are counted as participants if they attended at least one program class. For FUPTP, however, the available program data excluded youth attending fewer than 25 percent of the classes for which they were eligible in a given school year. Thus, the participation count shown for this program is a lower bound.
***p-value (of program-control difference) < 0.01; **p-value < 0.05; *p-value < 0.10, two-tailed test.
Estimated impacts on expectations are also more notable in size when focusing on participants only, though they are rarely statistically significant. For ReCapturing the Vision, program participants were 9 percentage points more likely to expect to abstain from sex through high school and 11 percentage points more likely to expect to abstain until marriage than their control group counterparts. However, neither difference is statistically significant (respective p-values = 0.34 and 0.13; not shown). For FUPTP, participants were 18 percentage points more likely than their control group counterparts to expect to abstain until marriage, a difference that is statistically significant (p-value = 0.04; not shown). Participants were also six percentage points less likely to expect to abstain through high school; however, this difference is not statistically significant.
 This measure of the mean age at first intercourse is based on the subsample of program and control group youth that reported having had sex. An alternative to this measure is the proportion of youth who report having had sex by a particular age (by age 14, for example). Regardless of the age cutoff examined, the findings indicate no statistically significant difference between program and control group youth.