This study compared changes in English speaking scores in two classes receiving Creative Strategy Learning (CSL) and discussion instruction at SMAN 1 Alas Barat. A quasi-experimental pretest–posttest design included two intact classes, identified as XI IPA 1 and XI IPA 2, with 25 students each. Oral picture-description tasks addressed grammar, vocabulary, pronunciation, comprehension, and fluency. The article recalculated the individual score columns because several gain entries and squared gains in the thesis were inconsistent. The CSL class mean increased from 58.88 to 70.72, while the discussion class mean increased from 53.52 to 57.36. Mean gains were 11.84 and 3.84 points respectively. The difference in gains was 8.00 points, with a 95% confidence interval of 6.00–10.00 and Welch’s t(35.60) = 8.12, p < .001, under student-level independence assumptions. The results indicate greater recorded improvement in the CSL class. However, one class per condition prevents separation of class and treatment effects. Baseline differences, limited assessment documentation, and uncertainty about implementation further restrict causal interpretation. Replication should include more classes, verified scoring procedures, and clearer records of instructional activities.
Copyrights © 2024