In this meta-analysis we explored whether Taekwondo practice has improved its participants’ physical self-concepts. We also tested the mediating influence of factors in past research such as country, participant age, and measurement properties in their associations with these taekwondo effects. We reviewed extensive data collected from Chinese, English and Korean participants in articles listed in Cnki, Wanfang, PubMed, Web of Science, KISS, RISS, and DBPIA databases. First, we evaluated the methodological quality of these published articles with Review Manager 5.4 software according to the Cochrane System Evaluation Manual. Then, we used Comprehensive Meta-Analysis 3.7 software for statistical analysis. We based these analyses on nine research studies containing a total of 1154 participants. We found a significant association between taekwondo activity and an improved body self-concept (ES = .688, p < .001). Subgroup analyses showed a stronger association between these variables in Korea (ES = .90, p < .001) than in China (ES = .34, p < .001), a stronger association for children (ES = 1.04, p < .001) compared to adults (ES = .46, p < .001), and a stronger association with the modified version of the Physical Self-Description Questionnaire (PSDQ) (ES = .99, p < .001) than with the original PSDQ (ES = .57, p < .001). We concluded that practicing Taekwondo led to improved physical self-concept, especially in a younger population. In addition, the experimental design and measurement methods may influence the apparent link between these variables.