Collaborative Research and Development Project. School Structure and Science Success: Organization and Leadership Influences On Student Achievement

学校组织和科学成就（SOSA）项目将记录解释不同学校招收不同种族和语言学生的科学成就差异的因素。研究的问题是：学校层面的哪些领导和组织特征与减轻科学成就差距有关？以往的学校效能研究表明，学校领导和社会资本对学生成绩有影响；SOSA项目的独特之处在于其对科学成就的关注。康涅狄格大学（University of Connecticut）和南佛罗里达大学圣彼得堡分校（University of South Florida St. Petersburg）的研究人员将与各自州的学区合作，确定学校领导实践可以与缩小与学生种族、英语流利程度和社会地位相关的成就差距联系起来。在这个为期五年的项目结束时，调查结果将以有关领导实践和学校组织的建议的形式出现，这些建议可以在其他学校环境中实施。本项目采用统计建模与定性数据相结合的混合方法设计。多元回归分析显示，那些五年级学生居住的学校在科学方面的成绩差距或大或小。运用社会资本理论（即学校规范、沟通渠道和可信度），通过对建筑校长、课堂教师和社区代表的访谈，对积极和消极离群学校进行比较。我们的期望是，与科学测试成绩在人口类别上存在差异的学校相比，为所有学生提供更公平的科学体验的学校将表现出更强的社会资本。这些见解将通过多层结构方程模型加以补充，以确定各种学校气候措施（例如，教师与校长之间的信任，教师与校长对领导的看法之间的对应关系，以及学校/社区关系）与通过全州五年级科学测试测量的科学成就之间的关联强度。此外，增长分析将用于检测随时间的变化，并提供有关政策变化或领导层调整之间联系的见解，因为科学成就差距受到影响。通过与两个州的150所学校合作，这个合作研究项目旨在产生适用于其他学校系统的发现。特别是在科学成就差距很大的环境中，特别是当学校之间的差距不同时，即使学生人数相似，这项研究的结果将具有实际的领导意义。该项目的专业知识包括科学教育、教育领导和统计建模。这种互补的组合增加了项目工作的深度，同时扩大了其潜在影响。该项目解决的关键问题包括：科学领域的领导力与其他学科领域的领导力在多大程度上相似或不同？领导设计的变化（例如，自上而下与分布式领导）如何有助于缩小科学成就差距？有效的领导能在多大程度上缓解其他因素，这些因素加剧了为每个孩子提供高质量科学学习体验的挑战？调查结果将与领导力发展战略一起通过SOSA项目网站发布。可交付成果包括复制研究的模板、专业发展的案例研究以及支持科学教师领导发展的策略。

The School Organization and Science Achievement (SOSA) Project will document factors explaining variations in science achievement across schools enrolling ethnically and linguistically diverse students. The research question is: what leadership and organizational features at the school level are associated with mitigating science achievement gaps? Previous school effectiveness studies demonstrate school leadership and social capital influencing student achievement; the SOSA project is unique with its focus on science achievement. Researchers at the University of Connecticut and the University of South Florida St. Petersburg, in collaboration with school districts in their respective states, will identify school leadership practices that can be connected with reductions in achievement gaps related to student ethnicity, English fluency, and social status. At the conclusion of the five-year project, the findings will take the form of recommendations about leadership practices and school organization that can be implemented in other school settings. The project uses a mixed methods design by combining statistical modeling and qualitative data. Multiple regression analyses highlight those schools populated by fifth graders that have greater or lesser achievement gaps in science. Using social capital theory (i.e., school norms, communication channels, and trustworthiness) comparisons of positive and negative outlier schools will be made via interviews of building principals, classroom teachers and community representatives. The expectation is that schools providing more equitable science experiences to all students will exhibit stronger social capital compared to buildings with disparities in science test scores across demographic categories. These insights will be supplemented by multilevel structural equation modeling to determine the strength of association between various school climate measures (e.g., teacher-to-principal trust, correspondence between teacher and principal perceptions of leadership, and school/community ties) and science achievement as measured by statewide fifth grade science tests. In addition, growth analyses will be used to detect shifts over time and provide insights about the links between policy changes or leadership adjustments, inasmuch as science achievement gaps are affected. By working with 150 schools in two states, this collaborative research project is designed to generate findings applicable in other school systems. Particularly in settings where science achievement gaps are large, and especially when such gaps vary between schools even when the student populations are similar, the findings from this study will have practical leadership implications. Expertise in this project includes science education, educational leadership, and statistical modeling. This complementary combination increases the depth of the project's efforts along with expanding its potential impacts. Key questions addressed by this project include: to what extent is leadership in science similar to or different from leadership in other subject areas? how do variations in leadership design (e.g., top-down versus distributed leadership) contribute to reductions in science achievement gaps? to what degree can effective leadership mitigate other factors that exacerbate the challenges of providing high quality science learning experiences for every child? Findings will be disseminated via the SOSA Project website, along with leadership development strategies. Deliverables include templates to replicate the study, case studies for professional development, and strategies for supporting the development of science teacher-leaders.