Synthesizing, Interpreting, and Extrapolating Interventions to Foster Human Development

其他项目信息-项目概要/摘要 "综合、解释和推断促进人类发展的干预措施" 将支持不同的儿童早期发展学者使用不同的实证策略来调查 成功的幼儿计划的治疗效果的来源，并比较技能的增长， 年龄跨越他们。它利用了三个有影响力的项目（佩里幼儿园）收集的大量数据 计划，初学者项目，牙买加达到和学习）和随后的实施和 适应我们的团队成员积极设计，收集和分析这些不同的子集， 程序通过随机分配进行评估。有些方案有长期的后续行动。这项研究有五个 主要目标：（1）使用共同的（跨研究）动态分析框架， 在不同学科中，我们分析了有关可比技能指标增长的纵向数据， 针对处境不利儿童的主要幼儿干预措施的成果， 和环境条件。我们进行调解和适度分析，以控制 干预引起的介质和调节剂的变化。我们研究了代内和代间 干预措施的影响与长期随访，使参与者进入中年（ABC）， 中年晚期（佩里）。这些分析调整了损耗和无应答。我们使用大样本 推理方法和基于设计的小样本随机化推理。我们协调措施， 结果，并要求我们使用的满足度量不变性标准;（2）建立在我们以前的研究， 我们使用经验验证的动态模型来预测样本外的实验结果。我们建立 有理论和实践基础的工具，用于估计短期干预措施的长期影响， （3）更新和扩展幼儿期的成本效益分析 程序.这是第一次，我们将对成年子女和兄弟姐妹的影响以及对 中年和中年后期的健康和收入的原始参与者。我们调查的影响 幼儿社会流动计划;（4）检查家访计划的有效性。我们 分析有影响力的牙买加达到和学习家访计划和许多计划的启发， 了我们将这些项目的儿童成长轨迹与更全面的综合项目进行比较 (and费用昂贵）的早期儿童方案，其中许多方案都有家访的组成部分。这将有利于 家庭访视在促进儿童发展中的作用的分离;（5）我们详细调查了家庭访视在促进儿童发展中的作用， 父母（照顾者）-儿童，家庭访客-儿童和家庭访客-照顾者的互动在促进技能 在牙买加Reach Up：China REACH（10个网站）的独特改编中，我们的研究成果得到了充分的记录。我们 定量评估"脚手架"对儿童技能发展的影响，每周进行一次测量。

OTHER PROJECT INFORMATION – Project Summary/Abstract “Synthesizing, Interpreting, and Extrapolating Interventions to Foster Human Development” will support diverse scholars of early childhood development using different empirical strategies to investigate the sources of treatment effects of successful early childhood programs and to compare the growth of skills by age across them. It draws on a wealth of data collected on three influential programs (the Perry Preschool Program, the Abecedarian Project, and Jamaica Reach Up and Learn) and subsequent implementations and adaptations. Members of our team have actively designed, collected, and analyzed different subsets of these programs evaluated by random assignment. Some programs have long-term follow-ups. This research has five main aims: (1) Using common (across studies) dynamic analytical frameworks developed and applied in different disciplines, we analyze longitudinal data on the growth of comparable measures of skills and outcomes from major early childhood interventions targeted to disadvantaged children, controlling for family and environmental conditions. We conduct mediation and moderation analyses to control for endogeneity of intervention-induced changes in mediators and moderators. We examine the intra- and inter-generational impacts of the interventions with long term follow up, taking participants well into their middle age (ABC) and late middle age (Perry). These analyses adjust for attrition and non-response. We use both large sample inferential methods and design-based small-sample randomization inference. We harmonize measures and outcomes and require those we use to satisfy metric invariance criteria; (2) Building on our previous research, we use empirically validated dynamic models to forecast experimental outcomes out-of-sample. We build theoretically- and empirically-grounded tools to estimate the long-run impacts of short-run interventions that will be available to other practitioners; (3) Update and extend cost-benefit analyses of early childhood programs. For the first time, we incorporate impacts on adult children and siblings as well as impacts on middle age and late middle age health and earnings of the original participants. We investigate the impact of early childhood programs on social mobility; (4) Examine the effectiveness of home visiting programs. We analyze the influential Jamaica Reach Up and Learn home visiting program and many programs inspired by it. We compare child growth trajectories from these programs with those from more comprehensive omnibus (and expensive) early childhood programs, many of which have a home visiting component. This will facilitate isolation of the role of home visiting in promoting child development; and (5) We investigate in detail the roles of parent (caregiver)-child, home visitor-child, and home visitor-caregiver interactions in promoting skill development in a uniquely well-documented adaptation of Jamaica Reach Up: China REACH (10 sites). We quantitatively assess the impact of “scaffolding” measured on a weekly basis on the growth of child skills.