AbstractThe problem in this research is whether the Problem Based Learningmodel of learning can have a better influence than the Direct Instructionlearning model, on the ability to service on volleyball, in class XI IslamicHigh School Pontianak Pontianak. In accordance with the objectives of thestudy, this study uses quasi-expremental experimental research. Pseudoexperiments are defined as experiments that have treatments, impactmeasurements, experimental units but do not use random assignments tocreate comparisons in order to conclude changes caused by treatmentrefinement or improvement of learning processes and practices. Thesubjects in this study were 62 students in grade XI of Pontianak CityIslamic High School. This research technique uses a lattice instrumentmeasuring instrument learning outcomes for volleyball. Data analysis usesquantitative data. The results of this study indicate that there aredifferences in learning outcomes of service on volleyball with problembased learning and Direct Instruction learning models in class XI studentsof Islamic Islamiyah in Pontianak. This is evidenced by an increase inservice skills for volleyball which is quite good, in the problem-basedlearning group, the results of calculations with an average post-test scoreof 31.96 and the direct instruction group of 28.69. Conclusions based onthe results of the research and discussion in the previous chapter, thegeneral conclusion in this study is that there are differences in learningoutcomes between the direct instruction learning model and problem basedlearning on the ability to service the XI grade students in PontianakIslamic High School in Pontianak.Keywords : , Comparison of Problem Based Learning, Direct Instruction, Learning Outcomes, Volleyball Service