Impact of ZEP QUIZ on Vocabulary Acquisition and Engagement among Year 3 Remedial ESL Learners: A Quasi-Experimental Study

Abstract

The persistent digital divide in Malaysia poses a significant challenge to educational equity, particularly for remedial learners residing in rural areas where exposure to the English language is minimal. This study empirically examines the efficacy of ZEP QUIZ, an AI-driven, gamified e-learning platform, in enhancing English vocabulary acquisition and learner engagement among Year 3 remedial students. Situated in a Chinese-medium primary school (SJKC) in the rural district of Dalat, Sarawak, the research addresses the chronic underperformance of students who lack environmental exposure to English. Adopting a quasi-experimental pre-test/post-test non-equivalent control group design, 12 remedial learners were purposively selected and matched by baseline proficiency. The Experimental Group (n=6) underwent a six-week intervention using ZEP QUIZ, while the Control Group (n=6) utilized traditional static worksheets. Data were triangulated using a validated Vocabulary Achievement Test (KR-20 = 0.82), a Student Engagement Survey adapted from Fredricks et al. (2004) (Cronbach’s α = 0.79), and semi-structured interviews. Non-parametric statistical analyses revealed that the Experimental Group achieved a statistically significant improvement in vocabulary scores (Z = -2.20, p = .028), with a large effect size (r = 0.63). In contrast, the Control Group showed only marginal gains (p = .046). Furthermore, thematic analysis identified "visual scaffolding," "gamified autonomy," and "safe failure environments" as key drivers of engagement. The findings validate the potential of adaptive AI tools to bridge proficiency gaps in resource-constrained settings, aligning with the aspirations of the Malaysia Education Blueprint (2013-2025).