This research aims to get an overview of the differences in the increase in cognitive abilities and retention of college students who get basic physics learning based on value use scientific methods and without use scientific methods, and get an overview of the relationship between retention and cognitive abilities of college students who get basic physics learning based on value use scientific methods. The research conducted use a quasi-experimental method with a randomised control group design pretest-posttest design. The population is all college students of the Physics Education Study Programme 2016-2017 odd semester at one of the Private Universities in Jakarta. Samples of two classes selected by cluster random sampling. The results showed that there was an increase in cognitive abilities and retention of college students in both classes. However, the magnitude of the increase in the experimental class was more significant than the control class. The normalised gain average score for cognitive abilities in the experimental class is 0.78 which meets the high criteria, and in the control class is 0.65 which meets the criteria of being. While the normalised gain average score for student retention resistance in the experimental class is 0.68 which meets the criteria of being, and the control class is 0.47 which meets the criteria of being. Based on the average difference test, at the 95% confidence level, the results of the study show that value-based basic physics learning using scientific methods can significantly improve cognitive abilities and college student retention resistance. In addition, the results of the study also showed that there was a strong and significant relationship between cognitive abilities and student retention resistance with a correlation value of 0.63.