<![CDATA[Despite growing evidence supporting game-based learning in EFL vocabulary instruction, long-term retention effects of Kahoot remain largely underexplored, as most existing studies are limited to immediate post-test measurements, leaving its longitudinal impact on vocabulary retention empirically unverified. This study aimed to investigate the effectiveness of Kahoot-based instruction on long-term vocabulary retention and students' perceptions toward Kahoot among Grade XI EFL students at SMAN 12 Enrekang, Indonesia. Employing a quantitative quasi-experimental design with a non-equivalent control group, 48 students were divided into an experimental group (n = 24) receiving Kahoot-based instruction and a control group (n = 24) following rote learning methods across six instructional sessions. Data were collected through a 20-item vocabulary test administered at four measurement points, namely pre-test, post-test, and two delayed post-tests, and a 10-item Likert-scale questionnaire administered to the experimental group. Mixed ANOVA results revealed a significant Time * Group interaction effect (F(2.503, 115.145) = 3.277, p = .031, = .067), with the experimental group demonstrating significantly superior retention at the second delayed post-test (F(1, 46) = 5.126, p = .028, = .100), approximately five weeks after the completion of instruction, retaining a mean score of 62.29 compared to 45.62 in the control group. Students reported very positive perceptions toward Kahoot (overall M = 4.32), with enthusiasm (M = 4.71) and enjoyable learning atmosphere (M = 4.67) as the most valued features. These findings indicate that Kahoot-based instruction is significantly more effective than rote learning in sustaining long-term vocabulary retention, with practical implications for EFL teachers to integrate retrieval-based and game-based tools as a strategic approach to vocabulary instruction.]]>
