Unmasking Risk Factors of Bullying Behaviors Among Adolescents in Schools Using Logistic Regression
Authors
Yixuan Yang
Share
Annotation
The bulling behaviors among adolescnets has become a serious issue in the United States. According to the U.S. Department of Education’s National Center for Education Statistics (NCES), one out of every five (20.2%) students report being bullied at school and 41% of students who reported being bullied at school indicated that they think the bullying would happen again.
In this research, we investigated possible risk factors for bullying behaviors at school among adolescents and identified the most significant positive and negative factors through logistic regression. We used the 2021 Adolescent Behaviors and Experiences Survey data with features ranging from demographic information to the adolescents’ family condition. The response variable is whether an adolescent has been bullied at school during the past 12 months.
After processing the dataset, we built a logistic regression model to predict whether an adolescent is likely to be bullied. By investigating the logistic regression coefficients, we found that parents’ attitude toward the adolescent, gender, race, and the adolescents’ relationship to people at school are all risk factors. Specifically, we found that female white adolescents are more likely to be bullied at school. The logistic regression model has achieved an AUROC score of 0.74, with 62.1% true positive rate (TPR) and 30.9% false positive rate (FPR). This predictive model is helpful for healthcare professionals to identify and reduce the risk for the adolescents that are prone to be bullied and thus developing mental health related issues.
Authors
Yixuan Yang
Share
References:
[1] National Center for Educational Statistics. (2019). Student reports of bullying: Results from the 2017 School Crime Supplement to the National Victimization Survey. US Department of Education. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2015056 .
[2] National Center for Educational Statistics. (2019). Student reports of bullying: Results from the 2017 School Crime Supplement to the National Victimization Survey. US Department of Education. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2015056 .
[3] Google. Classification: ROC Curve and AUC | Machine Learning Crash Course. Accessed November 25, 2021. https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc.
[4] Kozma, Laszlo. "k Nearest Neighbors algorithm (kNN)." Helsinki University of Technology (2008).