The purpose of this study was to apply worked example-based learning as an effort to improve the representation ability of junior high school students. This research is a quasi-experiment study with a non-equivalent control group design. The population in this study were all students of class VIII SMP Negeri 1 Talang Kelapa, Banyuasin, South of Sumatera, with a purposive sampling technique taken two classes as the sample. Samples on select from is student in class VIII.3 were used as experiment class, and class VIII.4 used as a control class. The instrument that used in this research is a mathematical representation test. The validation of this instrument was carried out by an expert review. Analysis the data using t-test with independent samples test. From the results of data analysis show the average final ability of students' mathematical representations in the experimental class was the same as the average final ability of students in the control class with Asymp.Sig (1-tailed) = 0.086. But, the average value of the post-test in the experimental class was 60, and the average value of the post-test in the control class was 53.13. Its meaning that the average test score of students' mathematical representation abilities in the experimental class was higher than the average ability test score mathematical representation of students in the control class