An Intelligent Ecosystem for Science Writing Instruction

点击翻译按钮获取中文摘要

The ability to express scientific ideas in both written and oral form is an important 21st century skill. Teachers, employers, and college faculty lament the inability of many high school graduates to write clearly. This deficit in writing is due in part because teachers do not have the time to provide appropriate, timely feedback to students on their writing. This project would help teachers help students achieve these skills through automating an effective feedback process, in ways that are customized to particular disciplines and local classroom needs, particularly in high needs districts. The project will contribute to knowledge about how students learn to write and how computer assisted systems can support this learning.This project will develop and test three tools: 1) Teaching resources organized as developmental trajectories for teachers to use (e.g. from more simple to more complex; with diagnostics and strategies for addressing particular challenges); 2) A teacher dashboard that uses Artificial Intelligence tools to provide timely formative assessment to teachers by highlighting problem areas in their students' writing and peer reviews; and 3) An online teacher resource exchange to rapidly grow the set of appropriate assignments that can be used with this approach, critically filtered by student performance metrics. The project builds on a current system called SWoRD, which supports student peer reviewing in many disciplines within and beyond science. Working with six lead teachers and larger set of pilot teachers, the project will develop a trajectory of effective writing assignments in Biology, Chemistry, and Physics. In year three, there will be a summative evaluation with 90 teachers.