The research carried out aims to see how the problem-based learning model with outdoor study influences students' science learning outcomes compared to conventional learning models. The population of this research was 98 students of SMPN 12 Tanjung Jabung Timur, while the sample was 26 students of class VII A and 25 students of class VII B, taken using a purposive sampling technique where class VII A was the experimental class and class VI B as a control class. This research is quantitative research using a quasi-experimental method with non-equivalent control group design. The data collection technique used in this research was a multiple choice test and analyzed using the t-test to see differences in learning outcomes with two different learning models. Based on the results of tests carried out with 20 multiple choice questions, researchers found that there were striking differences in learning outcomes where the average posttest score for experimental class students was 60 while the control class was 39.8, which was then analyzed through paired difference tests (t-test) for looked at the differences and obtained results of 0.000 < 0.05, which means Ha was accepted and H0 was rejected, which can be concluded that there are differences in student learning outcomes who use the problem-based learning model with outdoor study compared to the control class which in this case uses the conventional model. The findings obtained can strengthen evidence that the influence caused by the use of a problem-based learning model with outdoor study on the science learning outcomes of high school students is very good.