[Purpose] Since most junior golfers spend more time on golf than in school, it is to provide information that can improve players' performance by analyzing the potential average of their physical self-concept according to the average at-bats representing a junior golfer's performance. [Method] Among junior golfers in Gyeonggi-do, players who participated in golf game rules classes organized by the Korea Middle and High School Golf Federation in 2023 were asked to respond to a survey and analyzed using SPSS 23.0 and Amos 23.0 for Window. [Result] First, there was no difference in the measurement model between shape identity, measurement identity, intercept identity, and factor variance identity. Second, it was found that there was a difference in the latent mean of physical self-concept factors according to the average at-bats. [Conclusion] First, form identity, measurement identity, intercept identity, and factor variance identity were sequentially secured, and the construct equivalence of physical self-concept factors was verified. Second, there is a difference in the potential average of physical self-concept factors according to the average at-bats, suggesting that the scale of physical self-concept can be used equally for junior golfers. Third, when the difference in the potential average of each sub-factor of the physical self-concept was presented in order of effect size, it was found in the order of endurance/body overall, physical competence, flexibility, physical activity, appearance, health/self-esteem, and muscle strength.