University of Northern Colorado faculty to research how artificial intelligence can be used to identify how well students are learning
By Kelly Ragan
Thanks to a National Science Foundation grant, two University of Northern Colorado faculty members will be able to use new research methods to fine tune teaching in college-level environmental science courses.
According to a news release from the university, assistant professor Chelsie Romulo and Professor Steven Anderson will spend the next three years developing a machine-learning program that can better capture how well students grasp certain concepts they’re taught.
Traditionally, students show what they know through multiple-choice questions and short answer essays. Romulo and Anderson hope the computer program they create will help educators find more reliable patterns of understating.
The research could change how educators test students on environmental science. According to the release, Anderson worked to develop a similar program to gauge learning in geological sciences.
"Our work 15 years ago on the Geoscience Concept Inventory showed us that it was very difficult and time-consuming to make a multiple-choice test that was both statistically valid and reliable, but we were able to accomplish that by administering it to more than 4,000 students," Anderson said in the release.
The idea is to collaborate with other universities to apply artificial intelligence to the grading process to get a more accurate understanding of student learning.
About the Award
Project title: Improving Undergraduate Science, Technology, Engineering and Mathematics
Grant award: $1.077 million over three years
Funding agency: National Science Foundation
Of note: Researchers are collaborating with Automated Assessment of Constructed Response Lab at Michigan State University and the Science Education Resource Center at Carleton College in Minnesota to help evaluate the assessment tool and conduct outreach.