This research is based on the problems by science teachers, namely the daily test scores of students, more than 50% of students get scores below the minimum completeness criteria. The purpose of this study is (1) to reveal the presence or absence of differences in learning outcomes between students who are taught using a cooperative learning model type Group Investigation (GI) with students who are taught using conventional learning models, (2) reveal the presence or absence of scientific attitude differences, among students who are taught to use a cooperative learning model type of Group Investigation (GI) with students who are taught using conventional learning models. This research is an experimental study with the research Static Group Comparison Design. The subjects in this study were 8th grade students of the 11th Junior High School in Padang 2018/2019 in 64 students. The research instrument used was in the form of posttest for knowledge competencies, observation sheets for attitude and skills competencies.The hypothesis in this study was tested using the t ^ test. The results of the t-test are known that in the competency of knowledge t count 2.18> t table 2.04, the competency attitude t count 3.36> t table 2.04, while the competency skills t count 1.40 <t table. This shows that the hypothesis is accepted for knowledge and attitude competencies. So it can be concluded that the application of the Group Investigation learning model can improve student learning competencies in the material of Excretion Systems at Padang 11 Public Middle School with a confidence level of 0.975