In this review, we have examined the main components of peer tutoring in mathematics in primary education (1st to 6th grade) settings. Fifty studies were included in this article. The review was intended to examine trends in this research field regarding roles of the participants, their age (same-age vs cross-age), the methodological approach taken (quantitative, qualitative or mixed), the type of quantitative design, the analysed variables and organisational matters (number of participants, program duration, sessions per week and duration of the sessions). Main results were also included for each article and effect sizes were calculated when possible. It was found that 75% of the studies revealed medium to very large effect sizes. The average effect size was reported to be large to very large (Cohen's d = 0.89; standard deviation = 0.60). Limitations, recommendations for future research, implications for practice and conclusions are also presented.