Preventing Teen Pregnancy
...dance. The theory was clearly expressed and concise. The relationships that were studied were between student problem behaviors at entry and exit and student sociodemographic factors. There were four variables among the sociodemographic factors that were found to be significant to the likelihood of problem behaviors: student's gender, family composition(living with one versus two parents), racial/ethnic minority status, and grade in school. Another relationship that was studied was the one between program participation effects and differences between groups at entry. They wanted to know whether the fact that the control group was either more or less problematic than the Teen Outreach Group at entry had an effect on program effectiveness. A third relationship that was studied looked at whether or not the intensity of the program, measured by number of classroom sessions attended and the number of hours of volunteer service, was related to program outcomes. There are many studies that related to the topic of teen pregnancy and school failure. Two are reviews of the most rigorously evaluated programs targeting teen sexual behavior(Frost & Forrest, 1995; Philliber & Namerow, 1995) which have identified nine programs that target teen sexual behavior, one which provided contraceptive services to teens and had a clear effect in reducing pregnancy rates(Winter & Breckenmaker, 1991). Other quasi-experimental research have examined the effects of contraceptive distribution in programs were Zabin, 1992; Zabin, Hirsch, Smith, Streett, & Hardy, 1986; Zabin et al., 1988a, 1988b and another examination of a program different than contraceptive distribution was the Perry Preschool project, a comprehensive, developmentally focused study by Schweinhart, Barnes, & Weikart, 1993. Other studies referred to were ones that examined school failure and dropout(Felner, Brand, Adan, & Mulhall, 1993; Oakland, 1992), and the need for more multipronged and developmentally oriented programs dealing more comprehensively with young people(Carrera & Dempsey, 1988; Donovan & Jesser, 1985; Donovan, Jesser, & Costa, 1988; Schorr, 1988). Participants in the experiment included 342 students who participated in the Teen Outreach Program and 353 control students, who did not participate, from 25 random assignment sites nationwide from 1991-1995. All students were in the ninth through twelfth grades and were randomly assigned to the participant or control condition by picking names out of a hat or choosing every other name on an alphabetized list. Demographic characteristics were obtained showing that an ethnically and socioeconomically diverse group of students were sampled and that females were highly overrepresented. The population that the results could most likely be generalized to would be all ninth through twelfth grade students whose school participated in the Teen Outreach program because those were the schools randomly assigned in this study. There were two instruments used in the study. The first one was a brief self-report questionnaire that asked about their age, grade level in school, race, predominant household composition(one versus two parent), and parent's education levels. The second measure was a self-report questionnaire that assessed student's problem behaviors, such as if they had ever been pregnant or caused a pregnancy, had failed a course during the prior year at school, or had been suspended in the prior year at school. These measures' reliability and validity has been supported and found that when collected sensitively, self-report instruments are among the least biased means of assessing adolescent problem behaviors(Elliot, Huizinga, & Menard, 1989; Farrington, 1973; Patterson, & Stouthamer-Loeber, 1984). The procedure for implementing the program occurred as part of the regular school curriculum over a single school year, with participation usually occurring as part of a class taken for credit. Students were assessed during the first several weeks of the school year and then again at program exit pr the end of the school year. The questionnaires were distributed by Teen Outreach facilitators during an early Teen Outreach class or in study halls and other school settings for control students. There were many alternative theories for the results that the researchers accounted for and factored into the analyses. One for example was the concern that group differences in demographics and problem behaviors at entry may have influenced results in a way that inflated the apparent effects of program participation, such as it may have been easier to work with and achieve positive change in the less troubled Teen Outreach group than in more problematic control group. They classified the final 22 program sites into those in which the control group was more problematic than th...