One research study using educational data mining (EDM) as a source of evidence is titled "Using Educational Data Mining to Predict Students’ Academic Performance for Applying Early Interventions." It analyzed 300 undergraduate students' records from a Computer and Information Science College using six data mining methods to predict academic achievement. The study found key predictive features such as GPA in the first four semesters, number of failed courses, and grades in core courses. It showed that EDM can help identify students at risk early, predict honorary students, and support customized instructional interventions based on individual needs.

What Educational Data Mining Can Tell Us

Predict students' academic performance and identify at-risk students early.

Discover honorary or high-achieving students for targeted scholarships or opportunities.

Identify significant features influencing academic success (e.g., GPA, course grades).

Help customize teaching strategies based on student capabilities.

Guide interventions to improve learning outcomes and reduce dropout rates.

What Educational Data Mining Cannot Tell Us

EDM may not fully explain the underlying causes of student success or failure, such as emotional or social factors.

It may not capture qualitative or contextual aspects of the learning experience.

Results depend heavily on data quality; incomplete or biased data limit accuracy.

EDM requires technical expertise to interpret results correctly.

It may overlook non-academic factors like motivation, personal circumstances, or teaching quality.