Implementation of a Deep Learning Approach with the Project Based Learning Model in the Element of Fashion Style and Design Development

This study aims to examine the effects, improvements, and differences in the skills and work attitudes of students through the integration of a Deep Learning approach with the Project Based Learning PjBL model in the Style and Design Development element of the Fashion Design and Production subject at vocational high schools. The study employed a quasi-experimental method with a pretest posttest non-equivalent control group design. The participants consisted of 41 eleventh grade students from the Fashion Design Program at SMK Widya Praja Ungaran, including 21 students in the control class that implemented conventional PjBL and 20 students in the experimental class that implemented PjBL integrated with the Deep Learning approach. The research instruments consisted of a skills assessment rubric and a work attitude scale that had been validated by six experts using the Percentage of Agreement test. Data were analyzed using the N Gain test and the Independent Sample t Test. The findings indicate that the N Gain value of skills in the experimental class reached 65.4%, which falls within the moderate category and is higher than the control class with 53% in the less effective category. The N Gain value of work attitude in the experimental class reached 92% in the effective category, while the control class obtained 72% in the fairly effective category. The results of the t test show significant differences in skills t = 13.175, p = 0.000 < 0.05 and work attitudes t = 3.212, p = 0.003 < 0.05. These findings indicate that the integration of a Deep Learning approach within the PjBL model significantly improves student skills and work attitudes in vocational fashion design learning.