Problem Based Learning with a Deep Learning Approach and the Mathematical Connection Skills of Vocational High School Students

This study examines whether students’ mathematical connection abilities differ between learning through Problem Based Learning integrated with a Deep Learning approach and learning through Discovery Learning integrated with the same approach. The research population consisted of Grade X Automotive Engineering students at a vocational high school during the 2025/2026 academic year. The sample comprised 76 students from two experimental classes, with 38 students in each class, selected through cluster random sampling. After the treatments were completed, data were obtained using a test measuring students’ mathematical connection abilities. Data analysis included the Lilliefors normality test, F-test for homogeneity, and t-test for hypothesis testing. The Lilliefors test showed that both groups were normally distributed, with calculated values of 0,111 and 0,093, respectively, both lower than the critical value of 0,143. The homogeneity test showed homogeneous variances, with a calculated of 1,693, lower than the critical of 1,729. The t-test produced a calculated of 1,911, exceeding the critical of 1,665, indicating a significant difference between the groups. The analysis confirmed a significant difference in students’ mathematical connection abilities between the two groups, with those receiving Problem Based Learning integrated with a Deep Learning approach obtaining a higher mean score than those learning through Discovery Learning with the same approach.