The main purpose of this research is to investigate the relationship between class identification and self-esteem, and to figure out the effect of sports spectating on self-esteem. The first hypothesis is that class identification will enhance a level of self-esteem. Second, sports spectating will affect self-esteem. Third, game outcomes will moderate the relationships between team identification and self-esteem. An experimental design is conducted. Participants answer the Rosenberg Self-Esteem (RSE) scale (1965) before and after watching sports. The two scores are then compared. Experimental design arranges two groups of junior high school students to watch volleyball games in person. The experimental group has 134 students from the four classes in the semi final games. The control group has 53 students from two classes whose class teams are not in the semi final games. The results show that the relationships between class identification and self-esteem are positive. The first hypothesis is supported. For the control group, the difference in self-esteem before and after watching games is not significant. For the experimental group, when the outcomes of supported teams are big win or small loss, the difference in self-esteem before and after watching games is significant. However, when the outcome of the supported team is big loss, the self-esteem does not change. The second hypothesis is partly supported and the third hypothesis is supported. Schools may cultivate students’ ability in sport spectatorship, hold sport competitions, and increase students’ self-esteem through sports spectating.
