12,445 papers · continuously updated · last export: 10 Aug 2026livingmeta.ai
← Browse all papers
AI evidence extraction

Personalizing K–12 STEM Education through Technology-Enhanced Learning and Learning Analytics

Umar Bin Qushem · UTUPub (University of Turku) · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

5/10
Relevance
1/4
Quality (LMQS)
E
Evidence
0
Citations

Methodology & findings

Study design

Mixed-Methods Research (MMR) design divided into three phases: (1) synthesis phase, (2) intervention phase, and (3) reflection phase.

Sample

N = 790, 4 groups

Primary method

Exploratory data analysis and statistical methods. Specific statistical tests and software are not identified in the abstract.

Main result

The research demonstrates that "the study employs a Mixed-Methods Research (MMR) design divided into three phases: (1) a synthesis phase (2) an intervention phase; and (3) a reflection phase" with data collected from two interventions involving 720 primary/lower-secondary students using ViLLE-tool for arithmetic operations over 9 months and 70 upper-secondary students using VR-tool for Life and Evolution lessons over 5 weeks. The abstract indicates the work "contributes to the field by advancing theoretical understanding of personalized education, providing empirical evidence from authentic classroom settings, and demonstrating how adaptive learning technologies and LA can foster personalized learning in K-12 STEM."

Reports effect sizes.

Research paradigm

Mixed-methods (pragmatist)

Author conclusions

The authors conclude that their research "contributes to the field by advancing theoretical understanding of personalized education, providing empirical evidence from authentic classroom settings, and demonstrating how adaptive learning technologies and LA can foster personalized learning in K-12 STEM. The work addresses critical gaps in the literature by developing an empirically grounded, theory-informed adaptive and context-sensitive personalized learning interventions framework in addition to offering a methodological blueprint for future research in the Educational Sciences."

Risk of bias

Selection bias: participants self-selected into intervention groups; Attrition: not reported across 9-month and 5-week studies; Confounders: socioeconomic status, prior achievement not clearly controlled; Teacher effects: variation in implementation across classrooms; No control group mentioned for either intervention; Different study durations and contexts between the two interventions (9 months vs 5 weeks); No random assignment mentioned; Selection bias risk (two separate cohorts with different durations and tools may not be comparable); potential attrition bias (9-month intervention with 720 students); no mention of randomization or control conditions; funding source not disclosed.

Open questions raised

  • The research addresses critical gaps by developing an empirically grounded, theory-informed adaptive and context-sensitive personalized learning interventions framework. Future directions include offering a methodological blueprint for educational sciences research.
  • The authors identify that "integrated intervention approaches for achieving STEM-skills and enhancing motivation towards STEM in K-12 settings remain underexplored compared to Higher Education applications" and that "traditional classrooms often struggle to address individual student's learning needs in core subjects such as mathematics and life sciences."
  • The paper identifies that traditional classrooms struggle to address individual student learning needs in core subjects such as mathematics and life sciences. It notes that integrated intervention approaches for achieving STEM-skills and enhancing motivation remain underexplored in K-12 settings compared to Higher Education applications.
Data: not_statedCode: not_statedExtracted from: pdfAgreement 66%

Explore related topics

Doctoral TrainingMixed Methods

Related papers

← Browse all papers