<![CDATA[This study examined the effectiveness of game-based discussion, defined as structured speaking tasks that use role cards, evidence tokens, timed problem solving, and peer rotation to guide argumentative interaction, in enhancing critical speaking, confidence, and linguistic accuracy in EFL classrooms. Using a quasi-experimental design, 156 secondary students in East Java, Indonesia, were assigned by intact classes to experimental and comparison groups. Data were collected through pre- and post-speaking tests assessed with analytic rubrics, questionnaires, classroom observations, and semi-structured interviews. Results showed that the experimental group achieved large gains in critical speaking (M = 62.45 to 78.12, d = 1.98), confidence (M = 58.77 to 75.56, d = 1.87), and accuracy (M = 60.33 to 74.45, d = 1.83), with all gains statistically significant at p < .001 and stronger than those in conventional instruction. The novelty of the study lies in integrating cognitive, affective, and linguistic dimensions within one instructional model rather than treating speaking fluency or motivation as isolated outcomes. Lexical precision and syntactic control were still developing under spontaneous pressure, but guided feedback and peer scaffolding supported gradual improvement. The findings offer practical implications for EFL teachers and curriculum developers seeking interactive, measurable, and learner-centered speaking instruction.]]>
Copyrights © 2026
