Focal Interest
Focal Interest
AI Summary of Peer-Reviewed Research

This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. [See full disclosure ↓]

Publishing process signals: MODERATE — reflects the venue and review process. — venue and review process.

Adaptive formative assessment showed high estimation accuracy

A young woman with blonde hair wearing glasses and a black jacket sits at a desk in a modern computer lab, typing on a keyboard while viewing a technical blueprint or CAD design on a large monitor, with other students and computers visible in the background.
Research area:Computer ScienceStudent Assessment and FeedbackPsychometric Methodologies and Testing

What the study found

The study found that a formative adaptive assessment framework for engineering education achieved high estimation accuracy and satisfactory reliability for formative use across most learner profiles. It also found reduced precision at the extremes of the proficiency continuum and imbalances in item exposure.

Why the authors say this matters

The authors say the framework supports competency-oriented feedback, learning monitoring, and instructional interpretation within a curriculum-aligned assessment structure. They position adaptive assessment as a pedagogically grounded tool for formative learning support, instructional decision-making, and quality assurance in engineering education.

What the researchers tested

The researchers proposed and evaluated a formative adaptive assessment framework that combines an item response theory (IRT; a statistical model for estimating ability from test responses) based computerized adaptive testing engine with a Bayesian network-based diagnostic component. The system used dichotomous multiple-choice items aligned with engineering learning outcomes, with item selection adapting to learners' proficiency estimates and diagnostic modelling prioritizing under-assessed competencies.

What worked and what didn't

Item calibration used empirical data from 612 university students in computer science, and system performance was examined in a simulation involving 500 simulated learners. The results showed high estimation accuracy (r = 0.912) and satisfactory reliability for formative use across most learner profiles. Reduced precision at the extremes of the proficiency continuum and imbalances in item exposure were also observed.

What to keep in mind

The abstract says the main structural limits were related to item bank coverage and curriculum representation rather than the adaptive algorithms themselves. Limitations are otherwise not described in the available summary.

Key points

  • The framework combined IRT-based computerized adaptive testing with Bayesian network diagnostic modelling.
  • It was designed for competency-oriented feedback, learning monitoring, and instructional interpretation.
  • Performance testing used data from 612 university students in computer science and 500 simulated learners.
  • Results showed high estimation accuracy (r = 0.912) and satisfactory reliability for most learner profiles.
  • Reduced precision at the proficiency extremes and item exposure imbalances were observed.

Disclosure

Research title:
Adaptive formative assessment showed high estimation accuracy
Publication date:
2026-03-03
DOI:
10.3991/ijep.v16i1.60479
OpenAlex record:
View
AI provenance: AI provenance information is not available for this post.

More posts