Contrasting data mining with traditional statistics as the theoretical framework for understanding its benefits and drawbacks¶

Theory · All theories
Evidence · no claims cited

Description¶

The article's organizing framework compares data mining, a process of "systematically, and automatically or semi-automatically, uncovering patterns in data," with traditional statistics. Statistics confirms theory-nested statements using null hypotheses and random samples; data mining shares roots with exploratory data analysis, seeks immediately actionable prediction, and does not require a well-defined background theory. The review reports that several scholars' compare-and-contrast treatments "turned out to be an important theoretical framework through which to understand the purported benefits and drawbacks of data mining," and it structures the whole synthesis around this lens.

Design Implications¶

Context¶

Requirements¶

• Requires researcher understanding of the research context and dataset, which leading data mining frameworks make explicit as critical to effective data mining

Constraints¶

• The framework treats the contrast as a tradeoff between population generalization with few variables and detailed, specific-level information, not simple superiority

Target Learners¶

• education researchers considering data mining approaches

Target Learning Objectives¶

• understanding methodological tradeoffs of data mining versus traditional statistics in education research

Claims¶

Related Theories¶

Examples¶

-

Key Sources¶

• Iwatani, Emi. (2018). Overview of Data Mining's Potential Benefits and Limitations in Education Research. Practical Assessment, Research & Evaluation, 23(15). http://pareonline.net/getvn.asp?v=23&n=15