A Recommendation System For Nurturing Students' Sense Of Belonging

Nurturing a sense of belonging in a classroom can positively impact student attrition, especially for underrepresented groups. In this work, we develop and study the effectiveness of a recommendation system that aims to foster belonging by sharing challenges and possible solutions that other students have had while taking the same course. Our system uses sentence transformers and calculates the similarity between students’ reflections. We measure participating students’ sense of belonging before and after they view the top 5 challenges and solutions received from other students. Using a significance level of 0.05, we found a p-value of 0.0145, indicating that there is a significant increase in overall belonging values. Students also rated 61% of the solutions as useful. This work allows future students to also benefit from the experiences of those before them. By showing students that those before them also had similar challenges and overcame them, we can show students that they do, in fact, belong among their peers. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.