Development and Validation of an Algorithmic Randomization Model for a CEFR-Aligned English Proficiency Test in College Students
PASAA · 27.05.2026 · 73 (spc) · 166-201
DOI: 10.66947/pasaa.v73ispc.2068 ↗Kurz gefasst
Fachartikel in der Zeitschrift PASAA, online erschienen am 27.05.2026. Verfasst von Jamjumrat Deeprom. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Amid growing demands for standardized, scalable, and equitable English language assessments in higher education, institutions face challenges related to item exposure, content imbalance, and the resource-intensive nature of manual test construction. This study developed and validated an algorithm-based item randomization and test assembly model for constructing CEFR-aligned assessments from a pre-validated item bank. Grounded in Assessment Engineering and expert input, the model comprises five components—Listening, Vocabulary, Usage and Functional Language, Structure, and Reading—mapped to CEFR levels A2 to C1. Phase 1 involved focus group consultation with English language specialists to design a randomized test blueprint. Phase 2 assessed psychometric properties using confirmatory factor analysis and reliability testing with 300 undergraduates. The test demonstrated excellent model fit, high internal consistency (α = .806–.894), and a clear factorial structure, with Usage and Functional Language emerging as the strongest predictor of overall proficiency. The algorithm ensured thematic balance, avoided item repetition, and upheld difficulty calibration—overcoming common challenges in manual test construction. These results support the model’s feasibility and relevance as a scalable solution for modernizing English language assessment in higher education.
Zitieren
APA
Deeprom, J. (2026). Development and Validation of an Algorithmic Randomization Model for a CEFR-Aligned English Proficiency Test in College Students. PASAA, 73(spc), 166-201. https://doi.org/10.66947/pasaa.v73ispc.2068
Vancouver
Deeprom J. Development and Validation of an Algorithmic Randomization Model for a CEFR-Aligned English Proficiency Test in College Students. PASAA. 2026;73(spc):166-201. doi:10.66947/pasaa.v73ispc.2068
BibTeX
@article{Deeprom2026,
author = {Deeprom, Jamjumrat},
title = {Development and Validation of an Algorithmic Randomization Model for a CEFR-Aligned English Proficiency Test in College Students},
journal = {PASAA},
year = {2026},
volume = {73},
number = {spc},
pages = {166--201},
doi = {10.66947/pasaa.v73ispc.2068}
}
EUVIMED