Feasibility of a Social Network Preventive Intervention

描述(申请人提供)：在这个项目中，我们将开发一种创新的干预措施，以防止药物滥用和其他不健康的行为。高危学生有两个容易识别和与学校相关的特征--他们对学校没有强烈的依恋意识，他们没有积极的朋友。这些特征也与其他风险和保护因素密切相关。先前的研究表明，聚集高危青年只会增加他们的风险和越轨行为。我们提出了一项战略，将在低风险群体中分配高危学生，以最大限度地发挥他们与亲社会同龄人发展友谊的潜力。我们将开发一种社交网络干预策略，将高危学生插入到学生小组中，增加他们积极接受社会的机会。在这个R34项目中，我们将：1.完善一个社交网络调查，评估学生在年级内的朋友网络、与学校的联系、兴趣和价值观。2.开发计算机应用程序，使我们能够测试一种识别高危学生的方法。3.开发计算机应用程序，使我们能够测试创建低风险群体的替代方法。(至少有四个将被考虑--聚类最接近的朋友，使用Newman-Girvan算法分组，弱链接聚类，和高中心性种子。)4.开发计算机应用程序，使我们能够测试将高危学生归入低风险组的替代方法。(至少要考虑两项--匹配小组分数和匹配个人分数。)5.收集学生的前测数据。6.测试识别高危学生、创建低风险小组和插入高危学生的每种替代方案 将学生分为低风险组，比较每个组的效果，以确定构建社会网络干预组的最佳方法。7.开展干预试点研究。 我们将为一组学校提供干预小组名单，并鼓励他们尽可能多地召集小组，提供社交机会。第二组学校将作为延迟治疗的控制组。8.对学生进行后测，评估在高危学生之间建立积极友谊的干预措施的相对有效性 以及改善他们与学校的联系。八所中学的学生将在连续两个学年的春季完成调查。第一年收集的数据将用于 测试高危学生识别、低风险组形成和高危学生插入算法。四所中学将在下一学年采用团队作业干预。所有八所中学的学生将在第二年结束时再次接受调查。我们假设，与对照学校的高危学生相比，干预学校的高危学生将有更多的朋友，报告说他们与学校的联系更紧密，出勤率更高，逃课更少，学校不端行为更少，学习成绩也更好。如果成功，这种干预措施将在随后的随机对照研究试验中测试预防药物使用的有效性。

DESCRIPTION (provided by applicant): In this project we will develop an innovative intervention for preventing drug abuse and other unhealthy behaviors. At-risk students share two readily identifiable and school-relevant characteristics - they do not have a strong sense of attachment to school and they do not have positive friends. These characteristics are also strongly related to other risk and protective factors. Previous research demonstrates that congregating high-risk youth only increases their risk and deviance. We propose a strategy that will distribute at-risk students among low-risk groups in a way that maximizes the potential for them to develop friendships with prosocial peers. We will develop a social network intervention strategy that inserts at-risk students into small groups of students increasing their chances of positive social acceptance. In this R34 project we will: 1. Refine a social network survey that will assess students' network of within-grade friends, bonding to school, interests, and values. 2. Develop computer applications that will allow us to test a method for identifying at-risk students. 3. Develop computer applications that will allow us to test alternative methods for creating low-risk groups. (At least four are to be considered - Clustering Nearest Friends, Grouping using the Newman-Girvan algorithm, Weak-Link Clustering, and High-Centrality Seeding.) 4. Develop computer applications that will allow us to test alternative methods for inserting at-risk students into low-risk groups. (At least two are to be considered - Matching Group Scores and Matching Individual Scores.) 5. Collect pretest data from students. 6. Test each alternative for identifying at-risk students, creating low-risk groups, and inserting at- risk students into low-risk groups, comparing the effects of each in order to identify optimal methods for constructing social network intervention groups. 7. Conduct a pilot study of the intervention. We will provide one group of schools with intervention group lists and encourage them to convene groups as often as possible to provide opportunities for social interaction. A second group of schools will serve as delayed-treatment controls. 8. Posttest students and assess the relative effectiveness of the intervention for creating positive friendships among at-risk students and improving their bonding to school. Students in eight middle schools will complete surveys in the spring of two consecutive school years. Data collected during the first year will be used to test at-risk student identification, low-risk group formation, and at-risk student insertion algorithms. Four middle schools will adopt the team assignment intervention during the following school year. Students in all eight middle schools will again be surveyed at the end of the second year. We hypothesize that at-risk students in intervention schools will have more friends, will report increased bonding to school, and will have better attendance, less truancy, less school misbehavior, and improved academic performance compared to at-risk students in control schools. If successful, this intervention will be tested for efficacy of preventing drug use in subsequent randomized control research trials.