<![CDATA[The low level of short serve skills among high school students in badminton is a problem that needs to be addressed immediately, as this technique is a crucial foundation of the game. This study aims to determine the effect of implementing a deep learning instructional model on short serve skills in badminton. The study employed a quantitative method with a pre-experimental approach and a one-group pretest-posttest design. The sample consisted of 30 students selected using purposive sampling. The instrument, an observation sheet for short serve skills covering seven assessment aspects, was validated for content validity (CVI = 0.92) and item validity (calculated r > table r), and was found to be reliable (ICC = 0.894; Cronbach’s Alpha = 0.867). Data analysis utilized descriptive statistics, the Shapiro-Wilk normality test, Levene’s test for homogeneity of variances, and the paired-sample t-test. The results showed that the mean pretest score of 8.60 increased to 20.60 on the posttest (an increase of 12.00 points or 139.5%). All students improved from the “inadequate” or “very inadequate” categories to the “adequate” (60%) and “good” (40%) categories. The paired-sample t-test yielded a calculated t-value of -29.094 with a significance level of 0.000 (p < 0.05) and a Cohen’s d effect size of 5.31 (very large category), so H₀ was rejected and H₁ was accepted. It is concluded that the deep learning model has a significant effect on badminton short serve skills through learning that emphasizes conceptual understanding, active engagement, and reflection.]]>
