At-risk/Dropout/Stopout/Graduation Prediction

Kai et al. (2017) pdf

• Models predicting student retention in an online college program
• J48 decision trees achieved much lower Kappa and AUC for Black students than White students
• J48 decision trees achieved significantly lower Kappa but higher AUC for male students than female students
• JRip decision rules achieved almost identical Kappa and AUC for Black students and White students
• JRip decision trees achieved much lower Kappa and AUC for male students than female students

Hu and Rangwala (2020) pdf

• Models predicting if a college student will fail in a course
• Multiple cooperative classifier model (MCCM) model was the best at reducing bias, or discrimination against African-American students, while other models (particularly Logistic Regression and Rawlsian Fairness) performed far worse
• The level of bias was inconsistent across courses, with MCCM prediction showing the least bias for Psychology and the greatest bias for Computer Science
• Multiple cooperative classifier model (MCCM) model was the best at reducing bias, or discrimination against male students, performing particularly better for Psychology course.
• Other models (Logistic Regression and Rawlsian Fairness) performed far worse for male students, performing particularly worse in Computer Science and Electrical Engineering.

Anderson et al. (2019) pdf

• Models predicting six-year college graduation
• False negatives rates were greater for Latino students when Decision Tree and Random Forest yielded was used
• White students had higher false positive rates across all models, Decision Tree, SVM, Logistic Regression, Random Forest, and SGD
• False negatives rates were greater for male students than female students when SVM, Logistic Regression, and SGD were used

Christie et al. (2019) pdf

• Models predicting student's high school dropout
• The decision trees showed little difference in AUC among White, Black, Hispanic, Asian, American Indian and Alaska Native, and Native Hawaiian and Pacific Islander.
• The decision trees showed very minor differences in AUC between female and male students

Gardner, Brooks and Baker (2019) [pdf]

• Model predicting MOOC dropout, specifically through slicing analysis
• Some algorithms performed worse for female students than male students, particularly in courses with 45% or less male presence

Baker et al. (2020) [pdf]

• Model predicting student graduation and SAT scores for military-connected students
• For prediction of graduation, algorithms applying across population resulted an AUC of 0.60, degrading from their original performance of 70% or 71% to chance.
• For prediction of SAT scores, algorithms applying across population resulted in a Spearman's ρ of 0.42 and 0.44, degrading a third from their original performance to chance.

Kai et al. (2017) pdf

• Models predicting student retention in an online college program
• J-48 decision trees achieved much higher Kappa and AUC for students whose parents did not attend college than those whose parents did
• J-Rip decision rules achieved much higher Kappa and AUC for students whose parents did not attended college than those whose parents did

Yu et al. (2021) pdf

• Models predicting college dropout for students in residential and fully online program
• The models had worse true negative rates and recall for underrepresented minority (URM) students and for male students in residential and online programs, whether their status was included or not
• The model was less accurate for URM students studying in residential program.
• The model was worse for male students studying in online program in terms of true negative rates, recall and accuracy
• Models for first-generation residential students showed worse accuracy and true negative rate (i.e., predicting power of sophomore year persistence in college)
• Models for first-generation residential students showed significantly better recall (i.e., proportion of correctly identified dropouts) than online peers, whether their status was included or not
• Model performed with significantly lower accuracy and true negative rate for residential students with greater financial need than online counterparts, whether their status was included or not