<![CDATA[General Background Collaboration skills are essential competencies in contemporary education, particularly within problem-based and interactive learning environments. Specific Background Problem-Based Learning integrated with deep learning and differentiated approaches has been proposed to support meaningful engagement and teamwork among vocational students. Knowledge Gap However, empirical evidence on how differentiated PBL combined with deep learning systematically strengthens collaboration skills in formal vocational education remains limited. Aims This study aims to examine the application of differentiated PBL with deep learning to strengthen students’ collaboration skills through iterative classroom action cycles. Results The findings show a progressive increase in collaboration skills from 30.22% in the first cycle to 87.21% in the third cycle, accompanied by improved teaching quality from 46% to 87%, indicating a strong positive relationship between instructional quality and collaborative performance. Novelty The study introduces the integration of differentiation across material, process, and product within PBL and combines qualitative and quantitative observations in a vocational classroom context. Implications These results highlight the importance of adaptive instructional design, learning style mapping, and structured group interaction to foster inclusive, collaborative learning aligned with 21st-century competencies. Highlights• Progressive cycles demonstrate substantial growth in cooperative performance indicators• Adaptive grouping and varied learning products support inclusive participation dynamics• Instructional quality aligns consistently with improved teamwork engagement KeywordsCollaboration Skills; Problem Based Learning; Differentiated Learning; Deep Learning; Vocational Education]]>
