The purpose of this study was to determine the effectiveness of self-compassion training in controlling anger among junior high school male students in Yazd. The statistical population of the study comprised all junior high school boys in Yazd from among whom 150 students were selected by random multi-stage cluster sampling method. The Anger Control Questionnaire was administered to them and the 30 students who obtained the lowest scores were chosen and randomly assigned to two experimental (N=15) and control groups (N=15). The experimental group was treated with eight 70-minute self-compassion sessions; however, the control group received no intervention. To collect data, the State-Trait Anger Expression Inventory-2 (STAXI-2) (Spielberger, 2003) was utilized in pre- and post-tests for both experimental and control groups. The data were analyzed using analysis of covariance (ANCOVA). The findings showed that self-compassion training was effective in controlling students' anger (p<0.001). Furthermore, self-compassion training reduced the subscales of feeling angry, feeling like expressing anger physically, angry temperament, angry reaction, anger expression-in, and anger expression-out  (P<0.05). Nevertheless, self-compassion training had no significant impact on feeling like expressing anger verbally.