Collaborative Research: RealVAMS-Getting Real-World Value from Value Added Models

合作研究：RealVAMS——从增值模型中获取真实世界的价值

• 批准号: 1336265
• 负责人: Walter Stroup
• 金额: $ 8.99万
• 依托单位: University of Nebraska-Lincoln
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-10-01 至 2017-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1336265&HistoricalAwards=false
• 关键词: Collaborative; Research; RealVAMS; Getting; Real

Value-Added Models (VAMs) in education are increasingly being used to attempt to measure the impact of a teacher on student achievement in a way that takes into account students? and peers? prior performance and demographics. Data that typically inform a VAM rely largely on student achievement data from state testing systems. The RealVAMs project tests a multi-dimensional mechanism to include other measures of student success such as scores from different assessment instruments (SAT, ACT) and categorical responses such as whether or not a student graduated or persisted in a STEM field. The real world model using multi-dimensional data is being tested using a longitudinal database from a large school district. The project is creating an open source accessible program to facilitate the expanded measurement of VAMS across a larger scale. This project has national significance as the use of VAMs for teacher evaluation is expanding rapidly across the country in schools and districts. A major objection to the use of VAMs in teacher evaluation has been the over-reliance on the use of state testing data. Opponents of the use of VAMs argue that this reliance contributes to a narrowing of curriculum and instruction to only what can be tested on state tests. Not all subjects are tested, so the use of VAMs does not support the evidence that can be used in the evaluation of teachers in non-tested subjects or grade levels. This project informs the educational practice community of schools and districts and provides for more nuanced and richer information about the impact of teachers for policy decision making.

教育中的增值模型(VAM)正越来越多地被用来试图以一种考虑学生的方式来衡量教师对学生成绩的影响？同龄人呢？以前的表现和人口统计数据。通常通知VAM的数据在很大程度上依赖于来自州测试系统的学生成绩数据。RealVAMS项目测试了一种多维机制，以包括其他衡量学生成功的指标，例如来自不同评估工具(SAT、ACT)的分数，以及分类反应，例如学生是否毕业或坚持在STEM领域。使用多维数据的现实世界模型正在使用一个大型学区的纵向数据库进行测试。该项目正在创建一个开放源码的可访问程序，以促进VAMS在更大范围内的扩展测量。该项目具有国家意义，因为VAM用于教师评估的使用正在全国各地的学校和地区迅速扩大。在教师评估中使用VAM的一个主要反对意见是过度依赖国家测试数据的使用。反对使用VAM的人辩称，这种依赖导致课程和教学的范围缩小到只能在州考试中测试的内容。并不是所有的科目都经过测试，因此VAM的使用并不支持可用于评估未测试科目或年级水平的教师的证据。该项目向学校和地区的教育实践界提供信息，并为政策决策提供关于教师影响的更细微和更丰富的信息。