Abstract
In this study a child-teach-child method was applied to male underachievers in elementary school. Fifteen underachieving fifth and sixth graders who were weak in mathematics taught fifteen second graders, who were also weak in math, replacing their classroom teachers. The tutor-tutee pairs met three times a week for about 4 months. At the end of this period tutees had significantly improved in mathematics on Standard Achievement Tests; tutors had significantly improved in mathematics on Standard Achievement Tests and in the classroom; and tutors had showed significant improvement in overall achievement at school. No significant progress in mathematics in the classroom was found for tutees, and there was no significant improvement in the self-concept of tutors.