Objectives The aim of this review was to assess the effectiveness of music therapy on improving the mental health of offenders in correctional settings. Methods Multiple databases and journals were searched to identify randomized controlled trials and quasi-randomized controlled trials of music therapy for offenders in correctional settings. Results Five studies (n = 409; predominantly male) were included in random-effects meta-analyses. Music therapy was effective for promoting offenders’ self-esteem (Hedges’ g =0.55, p < 0.001) and social functioning (g =0.35, p <0.05). Effects on anxiety and depression depended on the number of sessions. For both outcomes, the studies with 20 or more sessions had larger effects than the study that had fewer than 20 sessions, and this difference was statistically significant (Q = 11.88, df = 1, p <0.001, anxiety; Q = 9.16, df = 1, p = 0.002, depression). No significant effects were found on behavior management or between different music therapy approaches. Conclusions Music therapy may be helpful for offenders to improve mental health. Future studies should consider including offenders of both genders, using larger sample sizes, and examining long-term effects. Publication bias may be an issue for this review given the small number of studies and the small sample sizes.