The purpose of this study was to compare the effects of peer-mediated versus teacher-directed reading intervention on the reading performance of high school sophomores. Participants (n = 57) from the lowest 25th percentile of their sophomore class in reading were assigned randomly to peer-mediated or teacher-directed intervention. Fifteen to sixteen 25-minute intervention sessions over 12 weeks included listening passage preview, retelling, and main idea questioning. Linear regression revealed statistically significant main effects of intervention on maze selection, with both treatment groups outperforming a nonequivalent control group (ES =.69-1.00). There were no statistically significant differences on oral reading fluency or maze selection between the peer-mediated and teacher-directed groups. Participating in peer-mediated intervention and receiving more minutes of instruction were significantly associated with higher performance on the district reading assessment. Implications for allocating resources to tiered intervention in secondary schools are discussed.