Principal Investigator: Joseph Krajcik
Project Collaborators: WestEd, University of Georgia, and University of Illinois-Chicago
Project Dates: September 1, 2021 - August 31, 2025
Award Amount: $889,999.00
Funder: U.S. National Science Foundation
The significant shortage of workers in STEM careers is partially due to ineffective instructional practice. Among these, the value of educational assessment practice has been increasingly questioned to the point that assessments are perceived as negative, not informative, and may even be detrimental to students.
This project aims to explore the effectiveness of a machine learning-based assessment system in helping middle school science teachers make instructional decisions using automatically generated student reports (AutoRs). The "Potential of Automatically Scored Three-Dimensional Assessment" (PASTA) project team will create tools aligned with the Next Generation Science Standards (NGSS) to study how teachers use these automatically scored assessments. These assessments evaluate students' scientific practices, understanding of cross-cutting concepts, and use of core disciplinary ideas, which help them understand phenomena or solve complex problems. The project will also develop support strategies to enhance instructional effectiveness.
The collaborative team will widely disseminate various products, such as 3D assessment scoring algorithms, AutoRs, PCKSs, and the corresponding professional development programs, and publications to facilitate 3D instruction and learning.
