Abstract:

Based on the data of CEPS (2014—2015), this paper empirically tested the impact of physical exercise on adolescents’ cognitive ability using OLS, QR and PSM. It is found that, first, according to OLS, the cognitive ability of adolescents who often participate in physical exercise is 0.149 higher than that of adolescents who do not often participate in physical exercise. When the pre-ability is controlled, the cognitive ability can be improved by 0.049 for every unit increase in the time that teenagers take part in physical exercise. Second, physical exercise has the strongest promoting effect on adolescents whose cognitive ability is in the middle, while the heterogeneity of physical exercise is characterized by “inverted V-shaped” curve by using the method of QR. Regular physical exercise could reduce the intra-group differences in cognitive ability at low and high scores by using the method of UQR. Third, school work is not the decisive factor for teenagers to participate in physical exercise, while watching TV, playing online games and other extracurricular activities constitute an extrusion relationship for teenagers to participate in physical exercise. Cultural capital, human capital and educational expectation are positively correlated with teenagers’ participation rate in physical exercise by using model Probit. Finally, the influence of physical exercise on cognitive ability and its average processing effect fall into the value range of [0.095, 0.117], while ignoring the selective deviation will result in high net effect of physical exercise by using PSM.

Key words: physical exercise, cognitive ability, teenagers, propensity score matching