The purposes of this study were to (a) investigate the efficacy of a core fraction intervention program on understanding and calculation skill and (b) isolate the effects of different forms of fraction word-problem (WP) intervention. At-risk fourth graders (n = 213) were randomly assigned to the school's business-as-usual program, or one of two variants of the core fraction intervention (each 12 weeks, 36 sessions). In each session of the two variants, 28 minutes were identical, focused mainly on the measurement interpretation of fractions. The other 7 minutes addressed multiplicative WPs versus additive WPs. Children were pre-/posttested on fraction understanding, calculations, and WPs. On understanding and calculations, both intervention conditions outperformed the control group. The effect of intervention versus control on released fraction items from the National Assessment of Education Progress was mediated by children's improvement in the measurement interpretation of fractions. On multiplicative WPs, multiplicative WP intervention was superior to the other conditions, but additive WP intervention and the control group performed comparably. On additive WPs, additive WP intervention was superior to multiplicative WP intervention, which was superior to control.