Predictive Model for High School Student Sense of Belonging

Presentation author(s)

Jamie Gabrielle Lepito ’20, Avon, Connecticut

Major: Computer Science
Minors: Journalism; Political Science

Coco Charles ’20, Madison, Connecticut

Major: Computer Science


High school is both a difficult and transformative time for adolescents. A sense of belonging is essential for a student’s well being. It can affect many aspects of their lives, from their mental health to their academic achievement. School administrators, unsure of which students are most likely to feel unwelcome, often struggle to address this problem.

In our research, we use machine learning to develop classification models for predicting a student’s feeling of belongingness at their high school. We use a decision tree and support vector machine to determine what attributes correlate with an individual’s sense of belonging. Our work is done in the Python coding language using data exploration libraries such as Pandas and Matplotlib.

We use data from the 2018 Dane County Youth Assessment Survey, which surveyed 16,289 students from 19 high schools throughout Dane County, Wisconsin. The survey is conducted every three years and consists of 113 questions asking students about a variety of topics ranging from family life to screen time, with the goal of bringing attention to student concerns and needs. This dataset has been used as insight throughout Dane County for policy decisions, parental awareness, school boards, and agencies designed to help youth. In addition, it has been the basis of various peer reviewed studies regarding LGBTQ+ students, bullying, and substance abuse.

After using the decision tree and support vector machine, we compare them to assess the performance of each. With these results, we determine which model is best suited to identify students who may feel socially isolated. We hope these findings can be used to improve student inclusivity in high schools.


Eyad Said

This site uses cookies to improve your experience. Read our Web Privacy Policy for more information.

Got it! ×