Clustering, a prevalent class of machine learning (ML) algorithms used in data mining and pattern-finding--has increasingly helped engineering education researchers and educators see and understand assessment patterns at scale. However, a challenge remains to make ML-enabled educational inferences that are useful and reliable for research or…

State education data systems help policymakers use data to evaluate the impact of their efforts to improve education. By disaggregating the data -- that is, breaking it out by different student subgroups -- policymakers can ensure that their efforts address the needs of students who have been traditionally underserved in educational settings. Yet…

Science instructors are increasingly incorporating teaching techniques that help students develop core competencies such as critical-thinking and communication skills. These core competencies are pillars of career readiness that prepare undergraduate students to successfully transition to continuing education or the workplace, whatever the field.…

The study used earliest available student data from a flagship university in the southeast United States to build data mining models like logistic regression with different variable selection methods, decision trees, and neural networks to explore important student characteristics associated with retention leading to graduation. The decision tree…

This study examines factors affecting students' performances in an Introductory Sociology course over five semesters. Employing simple and ordered logit regression models, the author explains final grades by focusing on individual demographic and educational characteristics that students bring into the classroom. The results show that a student's…

The authors investigated the results of the Educational Testing Service Major Field Test (ETS-MFT) administered to business majors at a U.S. state university. Longitudinal trends and cross-sectional differences are documented, including significant performance differences among students of different majors. Findings suggest that a cohort affect…

We use data from the 1990/1994 Beginning Post-Secondary Survey to determine whether the factors associated with long-term attrition from higher education differ for students who initially enrolled part-time as compared to for students who initially enrolled full-time. Using a two-stage sequential decision model to analyze the initial enrollment…

Data-mining technology's predictive modeling was applied to enhance the prediction of enrollment behaviors of admitted applicants at a large state university. (Contains 4 tables and 6 figures.)

Focusing on student retention and time to degree completion, this study illustrates how institutional researchers may benefit from the power of predictive analyses associated with data-mining tools. The following are appended: (1) Predictors; and (2) Variable Definitions. (Contains 5 figures.)