SMU Data Science Review
Abstract
In this paper, we present a framework to identify school-level factors within North Carolina public school administration’s control that have a positive impact on school performance. Public school administrators struggle to improve the academic performance of their schools, as the most influential factors determining overall school performance are outside of their scope of influence. We consider the current circumstances responsible for poor performance in North Carolina public schools and their implications for future academic improvement. Our framework utilizes an extreme gradient boosting model to predict school performance scores using only school-level features that administrators can impact. By varying the inputs, administrators can estimate the potential improvements to school performance scores. We find that the number of short-term suspensions per 100 students in a school year is the most important feature used to estimate school performance scores, followed by the school’s average daily attendance. Altering these features while holding all else constant is found to change school performance scores by just a few points. However, our framework creates an opportunity for schools to identify areas for change that may ultimately improve academic performance.
Recommended Citation
Leeson, Olivia; Bean, Kelly; and Drew, Jacob
(2018)
"Identifying Areas for Change: A Case Study on North Carolina State Public School Performance,"
SMU Data Science Review: Vol. 1:
No.
3, Article 4.
Available at:
https://scholar.smu.edu/datasciencereview/vol1/iss3/4
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
