•  
  •  
 

SMU Data Science Review

Abstract

In this paper, we established a regression model to identify factors that significantly influenced both positively and negatively school-level yearly teacher attrition rate within North Carolina public schools using Belk Endowment Educational Attainment Data Repository for North Carolina Public Schools, district-level North Carolina IRS income tax data [2], and county-level North Carolina crime data[3]. Frequent teacher turnover has been negatively affecting teacher retention, student academic achievement, and educational attainment. School administrators are struggling to reduce teacher attrition rate since teacher retention is important to students academic achievements. Our regression model utilized an eXtreme Gradient Boosting(XGBoost) Regression model to predict school-level teacher attrition rates using controllable school features that school administrators can affect, tax revenue features and public safety features. Out of 252 features reviewed, we found that school-level features impacted the most on predicting school- level yearly teacher attrition rate. The most important features were the school-level percentage of teachers with 0-3 years experiences, school- level short term suspensions per 100 students, school-level percentage of teachers that have reached proficient standard 1, school-level percentage of students who have economical disadvantages and school-level percent- age of teachers who are licensed. Our findings provide opportunities for educational administrations to recognize key factors that can influence teacher attrition rates in North Carolina public schools.

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Share

COinS