Study shows that AI support can help college students avoid failing in their STEM classes.

Researchers have discovered a novel approach to enhancing academic performance among college students pursuing studies in science, technology, engineering, and mathematics (STEM) fields.

A recently published study from the University of Nebraska-Lincoln highlighted that the implementation of artificial intelligence (AI) interventions resulted in a significant improvement in student achievement in STEM courses.

The study revealed that retention rates in undergraduate STEM majors have dropped below 50%, with graduation rates approximately 20% lower compared to non-STEM majors. Mohammad Hasan, a researcher specializing in big data and AI at UNL, noticed this concerning trend in his own STEM courses at the university, which has a student population of nearly 24,000.

Hasan expressed concern about the number of students seeking advice on improving their grades towards the end of the semester.

Hasan was motivated to develop an AI-based support system that could provide early feedback to students regarding their performance throughout the semester, guiding them on potential outcomes based on their study habits.

Collaborating with Bilal Khan, a former UNL researcher and current professor at Lehigh University, Hasan trained an AI model using data from 537 students in a computer science class from 2015 to 2018.

In the fall of 2019, the model was tested on 65 undergraduates taking the same course, with 32 receiving personalized success projections via automated emails at specific intervals during the semester. The rest received a message stating that a prediction could not be made.

The results showed a significant increase in pass rates for students who actively checked and responded to the AI-based forecasts compared to those who did not.

The successful outcomes of the study led to Hasan securing a $600,000 grant from the National Science Foundation to develop a smartphone app called Messages From a Future You, intended to further support STEM students.

The app aims to offer tailored advice and interventions, drawing on insights beyond grades to address various aspects of student life that may impact academic performance.

The app, co-designed by Hasan, Khan, and Neeta Kantamneni, seeks to engage students in activities that enhance motivation, peer collaboration, and academic support.

By collecting daily information from students, the AI model will evolve to provide personalized messages and interventions based on individual experiences and progress, akin to face-to-face counseling.

Messages From a Future You is envisioned as a companion for STEM students, offering guidance and support to tackle academic challenges. The app will even feature an avatar resembling the user to enhance engagement.

Hasan emphasized the app’s role in identifying the root causes of poor performance and providing targeted remedies to address issues like declining motivation and engagement among STEM students.

The app’s goal is to reignite students’ enthusiasm and sustain their motivation throughout challenging STEM courses, fostering a supportive environment for academic success.