The purpose of this study was to determine the effects of reciprocal peer teaching (RPT) on the music performance skills of fourth grade string students using quasi-experimental, between-subjects designs. Students from a fourth grade orchestra class (N = 36) were grouped into one of the three conditions: (1) a free interaction RPT group, (2) an identical group, receiving an expert model of interaction strategies prior to RPT, and (3) a control group receiving traditional, whole-class instruction. Students in the experimental groups worked in pairs over six weeks, after which all students completed a performance of French Folk Song for evaluation. Group mean comparisons indicated a significant difference in performance scores between the RPT-only and control groups, with the RPT-only group scoring significantly higher, raising questions regarding the ineffectiveness of the modeling intervention. Results prompt further questions regarding the efficacy of expert modeling during RPT. Music educators, particularly those who work with beginning instrumentalists, are encouraged to implement RPT as an instructional tool to help promote student skill development and pro-social classroom behaviors while being mindful of teacher behaviors that may interfere with the efficacy of the RPT process.