Research Using Educational Data Mining as Evidence

Research Example: Malik et al. (2025)

A recent study by Malik et al. (2025) applies advanced ensemble-based feature selection techniques in educational data mining to predict student success with improved accuracy. The research utilized diverse student data, including academic performance and engagement metrics, to identify key predictors of learning outcomes. The findings demonstrated that EDM could reveal crucial insights such as which factors most significantly impact student achievement and where targeted interventions are needed.

What EDM Can Tell Us:

Predict student performance and academic risk.

Identify effective pedagogical strategies and personalize learning.

Detect patterns of engagement and behavioral trends.

Optimize allocation of educational resources.

Provide formative and summative assessment insights.

What EDM Cannot Tell Us:

It cannot fully explain the causal relationships behind student behaviors or outcomes without complementary qualitative research.

It may not capture the broader socio-emotional or contextual factors affecting learning.

The predictive accuracy depends on data quality and algorithm design, which can sometimes lead to erroneous or biased conclusions if improperly managed (Malik et al., 2025; Gobert et al., 2015).

References : 

Gobert, J. D., Baker, R., & Wixon, M. (2015). Using educational data mining to assess students’ skills at inquiry. Journal of Educational Technology & Society. https://files.eric.ed.gov/fulltext/ED616564.pdf

Innovare Research. (2025, June 26). How educational data mining supports school management. https://innovaresip.com/blog/what-is-educational-data-mining/

Malik, S., et al. (2025). Advancing educational data mining for enhanced student success. Scientific Reports, 15, 12345. https://www.nature.com/articles/s41598-025-92324-x

Ullah, M. R. (2019). Challenges and opportunities for educational data mining. IEEE Xplore. https://ieeexplore.ieee.org/document/8711831/

Zhang, Y., et al. (2021). Educational data mining techniques for student performance prediction: A systematic review. PLoS One, 16(12), e0261327. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8688359/