Making Smart Choices from the Start

Joey Cox, M.A.

INTRODUCTION

Kings County, California, has excessively high rates of unplanned teen pregnancy and STD infection among teens, especially among low-income, Hispanic teens.  The goal of the SMART program is to reduce the incidence of teen pregnancy and STDs among youth participants by encouraging them to abstain from sexual activity and other high-risk behaviors through a comprehensive abstinence education program.  The areas targeted for intervention have been communities with the highest incidence of poverty, teen pregnancy and birth rates, and sexually transmitted infections. The SMART program currently serves male and female adolescents between the ages of 10 and 18. SMART program participants attend after-school programs for four to eight hours a week. The number of students enrolled in the SMART program since its inception ranges from n= 106 students in 2004/ 2005 to n= 175 students in 2007/ 2008. 

METHODS

Evaluation of the Kings County, California, SMART abstinence until marriage program is based upon a two-group quasi‐experimental research design diagrammed as follows (where “O” signifies an observation or data point and “X” the intervention):

Data points are fixed in relationship to the school calendar (pre-test in the Fall, and post-test in the Summer). Outcome evaluation of the SMART program began during its second year (2005).  At the conclusion of Year 4, evaluation data has been collected in three years (2005, 2006, and 2007).   During each program year evaluation staff administered the most recent version of the OAPP core questionnaire at pre and post-test to SMART program participants and a group of demographically matched youth who were not exposed to the SMART program.  Only students who provided assent and parental consent participated in the data collection process.

Because matched-pairs comparisons where both pre and post-test measures exist are statistically more desirable than an “independent samples” approach to outcomes assessment, only students who completed both measures are considered here. However, one of the primary limitations of the evaluation stems from the fact that not all participating students completed both the pre and post-test survey.  Consequently, students with both pre and post-test across multiple project years were aggregated to maximize sample size.

Five hypotheses are tested to assess the impact of the SMART program:

1. An increase in the proportion of students in the SMART program who state their intention to remain sexually abstinent will be observed between pre and post-test. This change will not be observed among those in the comparison group.
2. Students in the SMART program will demonstrate an increased understanding of the relationship between alcohol and drug use and sexual activity between pre and post-test. This change will not be noted among those in the comparison condition.
3. Between pre and post-test, students in the SMART program will demonstrate an increased understanding of the risks of premarital sex. This difference is not expected among students in the comparison group.
4. At post-test a greater proportion of students in the SMART program will have set realistic goals for their future than at pre-test; a difference that is not expected among students in the comparison group.
5. Improved communication between parent and child regarding delaying sexual activity, risk avoidance, decision making, setting goals and health in general will be observed between pre and post-test among students in the SMART program. Again, these improvements will not be observed among students in the comparison group. 

RESULTS

Study Sample

The study sample consists of 199 students for whom both pre and post-test measures were available.   One hundred thirty-five (67.8%) were SMART program participants and 64 (32.2%) were in the comparison condition.  A higher proportion of females (n= 112; 56.9%) than males (n= 85; 43.1%) are represented in the study sample. At pre-test, the age of students ranged from nine to 17, with a mean of 11.76 years.  Nearly nine in ten students in the survey sample self identify as Latino/Hispanic (n= 177; 89.4%).  More than half of students represented in the sample lived with both parents at baseline (n= 107; 55.4%), 18.7 %(n= 36) lived in single parent households, and one quarter (n= 50; 25.9%) lived with another adult guardian.  With one exception, students in the SMART program and comparison group were similar with regards to the demographic characteristics described above. A significantly higher proportion of students in the SMART program were between the ages of 10 and 12 at pre-test (n= 101; 74.8%) than those in the comparison group (n= 36; 57.1%).

Hypotheses #1

The intent to remain abstinent was assessed using the following item: “When do you think you will first have sex?” Response options included 1) until I graduate from high school, 2) until I am at least 21, 3) until I am married, 4) I have already had sex and 5) I don’t know.  This item was only administered to students over the age of 18 in Years 2 and 3 of the program evaluation therefore the sample size is limited.  Moreover many students refused to answer this item. In total, only 30 students (12 in the SMART program and 18 in the comparison condition) provided a response to this item at both pre and post-test.  Using the Wilcoxon Signed Rank Test to test the study hypothesis; no significant differences were found.  

Hypotheses #2

Knowledge of the relationship between alcohol and sex was assessed by aggregating the responses to three items. Students rated their agreement with the following three items on a scale from one to four with 1=strongly disagree to 4=strongly agree; “It is not a big deal if people my age make decisions about sex when they are drinking or using drugs,” “Alcohol or drugs can influence your decision to do something sexual,” and “Doing something sexual while using alcohol or drugs can increase the risk of exposure to STDs and pregnancy.” Reverse coding the first item, the three items had an internal consistency of α =.618 at baseline and α =.484 at follow-up. Using the difference between the observed score at post-test and pre-test (where positive integers reflect change in the positive direction), a modest gain in knowledge of the relationship between sex and the use of alcohol and drugs is noted among youth between the ages of 10 and 12 in the SMART program (mean difference = .514). Although not statistically significant, this change is not observed among 10 to 12 years olds in the comparison group.  Additionally, while a decrease in knowledge is observed among 13 to 18 year olds in the comparison group (mean difference= -.1905), none is noted among SMART program participants of the same age.  

Results of the remaining three hypotheses are forthcoming. 

DISCUSSION

At the time of poster production, tests of the remaining three hypotheses are underway. However, tests of the first two hypotheses reveal small but notable benefits associated with participation in the SMART program, especially among younger students. The results of the outcome evaluation will be discussed in the context of process evaluation considerations.  For example, using the limited data available, the impact of attendance on project outcomes will be considered.

IMPLICATIONS

Implications will be developed pursuant to the final analysis of outcome data.

Contact

Social Science Research, CSU Fullerton
1222 West Lacey Boulevard, Hanford, CA 93230
559 583-0315
Fax: 559 583-0603
E-mail: jcox@kcao.org