From Smartphone to Smart Learning: A Quasi-Experimental Study of an AI-Enabled Mobile Application for English Vocabulary Learning
Mobile vocabulary applications may support out-of-class learning, yet access alone does not resolve irregular review, weak personalization, or limited monitoring of practice. This 12-week quasi-experimental teaching case compared a bundled AI-enabled vocabulary application with non-adaptive printed and digital word lists among 100 second-year non-English-major students in eight intact tutorial classes at a university in eastern China. Four matched class pairs were randomized at the class level, and complete-case analyses included 46 students per condition. Vocabulary performance was assessed with a 100-point same-item test administered in counterbalanced orders; application logs, comparison-condition study diaries, attendance records, and a four-dimensional attitude questionnaire were collected. A linear mixed model with a random intercept for class estimated a 6.6-point adjusted posttest advantage for the application condition, 95% CI [1.3, 11.9], p = .022, with an intraclass correlation of .06. A conventional student-level ANCOVA yielded a 7.0-point difference, F(1, 89) = 18.42, p < .001, partial eta squared = .17. Active study days were associated with vocabulary gain in the unadjusted analysis, r = .46, p = .001, and remained associated after adjustment for pretest score and course attendance, B = 0.21, p = .014; the adjusted association for review completion was weaker. Perceived usefulness received the highest rating (M = 4.12), whereas continuance intention was lower (M = 3.61). The findings illustrate a plausible short-term advantage for the complete instructional package, while the small number of classes, same-item testing, self-reported comparison-condition study time, and bundled intervention prevent feature-specific causal conclusions.
Authors
- Mohamad Jafre Zainol Abidin
- Liyun Chen
Institutions
- Malaysia University of Science and Technology (MY)
Publication Details
- Journal
- Language Ecology and Application
- Published
- 2026-11-20
- DOI
- https://doi.org/10.63808/lea.v2i1.471
- Primary Topic
- Mobile Learning in Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00