A machine learning approach to prioritizing students at risk of not graduating high school on time

Type: Evidence | Proposition: C: Uptake | Polarity: | Sector: | Country:

The paper deals with two key ideas that could help schools graduate more students on time. The first is to produce a ranked list that orders students according to their risk of not graduating on time. The second is to predict when they'll go off track, to help schools plan the urgency of the interventions. Both of these predictions are useful in identification and prioritization of students at risk and enable schools to target interventions. The eventual goal of these efforts is to focus the limited resources of schools to increase graduation rates.

The results of this study have helped a school district systematically to adjust analytical methods as they continue to build a universal EWI (early-warning indicator) system. The district is also highly interested in the web-based dashboard application that was developed.

Citation: Aguiar, Everaldo, Lakkaraju, Himabindu, Bhanpuri, Nasir, Miller, David, Yuhas, Ben, & Addison, Kecia L. (2015). Who, when, and why: a machine learning approach to prioritizing students at risk of not graduating high school on time. Paper presented at the Proceedings of the Fifth International Conference on Learning Analytics And Knowledge. | Url: http://d-miller.github.io/assets/AguiarEtAl2015.pdf

Leave a Reply