School Punishment, Special Education, and Childhood Behavior Problems

在美国，公众和学术界越来越关注学校纪律和特殊教育政策的变化。一些研究人员认为，学校的纪律政策变得越来越严厉，导致越来越多的学生在各个年级被停学。与此同时，随着学校扩大特殊教育计划，为医学上公认的行为障碍儿童提供服务，越来越多的儿童正在接受ADHD等疾病的诊断和治疗。学校纪律和特殊教育的做法是为了让成年人认识到和应对感知儿童的不当行为。因此，学校惩罚、特殊教育以及行为障碍的诊断或治疗的经历应该是适当的，并能改善未来的行为。然而，人们日益关切的是，旨在改善所有青年学习环境的学校纪律和特殊教育做法和干预措施在不同种族和族裔背景的儿童中的适用情况并不均衡。非西班牙裔白色儿童更有可能在有健全的特殊教育计划的学校上学，并且比种族和少数民族儿童的学校惩罚率更低。因此，少数种族和族裔儿童的停学率高于白色儿童，他们接受高质量特殊教育或医疗的可能性也较低。随着证据继续证明早期学校惩罚对学术和行为的有害影响，越来越多的学校推动采取考虑潜在行为问题和障碍的方法来处理感知的不当行为。了解学校纪律，特殊教育和行为问题的诊断/治疗如何影响所有儿童，包括边缘化和高危青少年，将有助于制定最适合美国行为问题儿童的政策。 关于学校纪律和特殊教育是促进还是破坏教育进步，存在着分歧。此外，很少有研究认为学校惩罚和特殊教育或治疗/药物是对儿童社会控制的对立或合作方法。该项目结合了来自美国大城市风险家庭样本的数据（脆弱家庭儿童福利研究）以及地区和学校一级关于纪律和特殊教育做法的数据（例如民权数据收集）。该项目使用两种广泛的统计方法来回答多个问题。首先，标准的线性和非线性回归模型和匹配模型（例如倾向得分加权）将用于评估学校纪律和特殊教育政策和做法如何影响（1）成人对儿童行为的看法;以及（2）儿童的停学和/诊断/治疗经历。这将通过比较在不同停学率和特殊教育入学率的学校上学的儿童来实现。第二，调解和温和的模式，包括互动条款，将被用来检查是否照顾者和/或教师的儿童行为的看法解释学校纪律和特殊教育的做法，儿童社会控制的经验，和未来的行为问题和学业成绩之间的关系。该项目将通过使用匹配技术，固定效应估计和安慰剂回归来解决调解和缓和模型中的选择问题。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

There is growing public and scholarly attention given to changes in school discipline and special education policies in the United States. Some researchers suggest that school discipline policies have become increasingly severe, resulting in a growing number of students receiving school suspensions at all grade levels. At the same time, as schools expand special education plans to provide services for children with medically recognized behavior disorders, a growing number of children are receiving diagnoses and treatment for conditions such as ADHD. School disciplinary and special education practices are intended to allow adults to recognize and respond to perceived child misbehavior. As such, experiences with school punishment, special education, and diagnoses or treatment for behavior disorders should be appropriate and improve future behavior. However, there is growing concern that school discipline and special education practices and interventions intended to improve the learning environment for all youth are instead applied unevenly across children from different racial and ethnic backgrounds. NonHispanic White children are more likely to attend schools with robust special education programs and lower rates of school punishment than racial and ethnic minority children. Consequently, racial and ethnic minority children have higher rates of suspension and are less likely to receive quality special education or medical treatment than White children. As evidence continues to demonstrate the harmful effects of early school punishment on academics and behavior, there is a growing push for schools to adopt approaches to perceived misbehavior that consider underlying behavior problems and disorders. Understanding how school discipline, special education, and diagnoses/treatment for behavior problems affect all children, including marginalized and at-risk youth, will help to craft policies that best serve children with behavior problems in the United States. There is disagreement regarding whether school discipline and special education improve or disrupt educational progress. Additionally, few studies consider school punishment and special education or therapy/medication as opposing or collaborative approaches to child social control. This project combines data from a sample of at-risk households in large U.S. cities (Fragile Families Child Wellbeing Study) and district- and school-level data on disciplinary and special education practices (e.g. Civil Rights Data Collection). The project uses two broad statistical approaches to answer multiple questions. First, standard linear and non-linear regression models and matching models (e.g. propensity score weighting) will be used to estimate how school discipline and special education policies and practices affect (1) adult perceptions of child behavior; and (2) child experiences of school suspension and/ diagnoses/treatment. This will be achieved by comparing children attending schools with different suspension and special education enrollment rates. Second, mediation and moderation models, including interaction terms, will be used to examine whether caregiver and/or teacher perceptions of child behavior explain the relationships between school discipline and special education practices, childhood social control experiences, and future behavior problems and academic achievement. The project will address issues of selection in mediation and moderation models by using matching techniques, fixed-effects estimation, and placebo regression.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.