Evaluation Plan For Trail, An Abstinence-Based Adolescent Pregnancy Prevention Program

Janice K. Janken, PhD, RN, Queens University of Charlotte,
Barbara K. Sheppard, MSW and Megan Canady, BS, Cabarrus Health Alliance, Kannapolis, North Carolina

Introduction and Program Description

TRAIL (Taking Responsible Actions In Life) is a 3-year, four component intervention designed for students in grades 7, 8, and 9 and their parents. TRAIL seeks to increase the knowledge and skills needed to remain abstinent, change the perception of students to acknowledge that abstinence is the normal behavior among peers, and to increase parent-child connectedness as a protective factor against early sexual activity and other risky behaviors. TRAIL is a saturation model designed to target all students and parents at the intervention school-sites. The four program components are: 1) Abstinence education through which students receive fifteen one-hour in-school lessons/year; 2) A social norms marketing campaign in which monthly topics such as relationships with parents, or alcohol and drug abuse are expanded upon through bulletin boards, newsletters, and an associated community service project, 3) Sponsorship of youth development activities that include mentoring from older students, after-school tutoring and service learning projects, and summer participation in community events; and 4) Encouragement of parental involvement through activities that advance parent-child interaction such as homework, monthly parent newsletters, and family events.

The questions to evaluate if the 3-year TRAIL intervention is effective are:

Between 7^th, 8^th, and 9^th graders who participate in TRAIL and those who do not, are there statistically significant differences in:

family communication patterns, extra-curricular involvement, decision-making about risk behaviors, self-confidence, future orientation, and attitudes about abstinence;
perceived developmental asset strengths;
the congruence of perceived and actual social norms toward healthy behaviors; and
parents’ perceived communication patterns in discussing sexually related issues with their children?

Methods

Design:

The evaluation design is a two group repeated measures quasi-experiment. Students are assigned on the basis of school attended to either the intervention or comparison group. The intervention lasts three years, but dosage of the intervention is contingent on students’ grade. Students beginning the intervention in 7^th grade will receive all three years of the program, students beginning in 8^th grade will receive two years and students beginning in 9^th grade will receive one year.

Sample:

The intervention is based in Cabarrus County, North Carolina in which there are two separate school systems. The intervention sample consists of students attending a middle school (7^th and 8^th grade students) and the associated high school (9^th grade students) in the Kannapolis school system and their parents. The control sample consists of students and parents from a middle and high school in the Concord school system. The controls schools are similar in ethnic and economic characteristics and size to the intervention schools. Only students with written parental consent and student assent are included in the evaluation. Approximately 4,200 students will be eligible to participate, 2,150 in the intervention group and 2,150 in the comparison group.

Data Collection Instruments:

1. AFL Core Baseline and Follow-up questionnaires to assess communication patterns, extra-curricular involvement, decision-making about risk behaviors, self-confidence, future orientation, and attitudes about abstinence.

2. Developmental Assets Profile to assess perceived internal and external asset strength.

3. Youth Norms Scale to assess actual personal behavior, attitudes toward the behavior, and perceptions of peers’ behavior and attitudes.

4. Parent Questionnaire to assess parent-child communication regarding sexual issues.

Data Analysis Plan: The intervention and control groups will be compared to test for statistically significant differences in measurements of:

family communication patterns, extra-curricular involvement, decision-making about risk behaviors, self-confidence, future orientation, and attitudes about abstinence,
perceived developmental asset strengths,
the congruence of perceived and actual social norms toward healthy behaviors, and
in parents’ perceived communication patterns in discussing sexually related issues with their children

The main analysis will compare students receiving the full, 3-year intervention (n≈350 in each group). Dose-response will also be assessed for students receiving one or two years of the intervention. Statistical tests will depend on the variable’s level of measurement (e.g. Chi-square, Mann-Whitney U, or t-test).

Results

The 1^st year of the project has been completed and was used as a pilot to:

trial the intervention and evaluation procedures at a school different from those that will participate in project evaluation, and
work with the schools that will participate in the program evaluation to determine how best to collect parental consent, administer surveys, and provide incentives for participation.

The trial of intervention components was conducted with 700+ students, 10 teachers, 120 parents, 35 community members, and 29 older student mentors. Pilot study data suggest that the process of delivering the intervention and evaluation procedures will be effective. Length of time needed to complete data collection was established and three versions of the Social Norms survey were created and pilot-tested to determine which was perceived most favorable by students. A major challenge to the evaluation anticipated will be subject retention since the project intervention is three years and requires longitudinal participation.

Discussion

The pilot of the evaluation influenced future data collection protocols and procedures. Evaluation staff learned the necessity to work with each individual school to create an implementation evaluation plan, the importance of showing parents that this is a school supported evaluation, the need to include teachers in the distribution and collection of consent forms to increase participation, and to provide incentives that students deem as valuable for survey completion. This pilot year provided invaluable data on both the program implementation and evaluation process for the intervention.

Implications

This project evaluation has the potential to contribute to the study of the effect of school-wide multi-component services for multiple years in the effectiveness to promote abstinence. Second, this program could further contribute to the research regarding the importance of parent-child connectedness to prevent early sexual activity. Third, limited research exists regarding the impact of social norms marketing on sexual decision making of youth, and this demonstration project examines this factor. Possible implications of this demonstration program are that adolescents are more likely to remain abstinent when they are given multiple years of intervention programming. A possible implication to school-based evaluation is the importance of involving school teachers and administration in developing the best way to implement the evaluation at each school-site.

Contact
Janice K. Janken, PhD, RN
Telephone: 704-337-2382
E-Mail Address: jankenj@queens.edu