In this study McDonald used educational data mining to predict the retention status of future students identifying with certain minority groups at Middle Tennessee State University. The researcher hoped to determine if there were particular attributes that were predictive of retention. The four minority groups identified for the purpose of the study were:

1. African American Students
2. First Generation Students
3. Disabled Students
4. Hispanic students

In general, data mining is used to uncover different data patterns that may be hidden in large sets of data. Cluster analysis is one approach of data mining used to detect interesting patterns in the data. Another approach is anomaly detection which looks for irregularities or unusual groupings in the data. Classification is used to find the correlational relationships between large sets of data and make predictions based on those relationships. Classification is the primary method of data mining used in this study.

The attributes found to be the best predictors of student success were the istitutional GPA and the students' financial situations. Other predictors explored were types of math courses taken in high school and college, number of courses from whom the student withdrew, and parents education level.

The findings of this study show that first generation students are retained at a higher rate when the student has a high GPA ( >= 2.75 ). In contrast, first generation students with low GPAs had the highest drop rate. African American and Hispanic students with lower GPAs have the same retention rates as students in othe minority groups with similar GPAs. Of all of the minority groups examined, disabled students have the highest drop and transfer rates, meaning these students are retained less. 

This is an example of how data mining can be used to increase equity and diversity on college campuses. When students with certain attributes are identified, more supports and intervetions can be put in place to help retain these students, thereby increasing or maintaining the diversity on campus.

McDonald, Kailey (2015) Mining Educational Data to Create a Model to Predict Student Retention (Doctoral Thesis).