This article reports on 2 experiments in inner-city Baltimore evaluating a computer-assisted tutoring approach, Tutoring With Alphie (TWA), in which 1 paraprofessional can work with up to 6 children at a time. In Study 1, we randomly assigned 14 schools to receive TWA or to continue with whatever approaches they were currently using. Each experimental school (n = 8) received a half-time paraprofessional tutor. Struggling readers in the lowest 30% of Grades 1-3 received tutoring using TWA. In comparison to control schools (n = 6), reading outcomes strongly favored TWA (effect size = +0.46, p < 0.01). In Study 2, new students in 7 of the 8 TWA schools received tutoring, and 6 schools continued as controls. Results again favored the TWA group (effect size = +0.40, p < 0.001). The findings support the effectiveness and cost-effectiveness of using technology to offer tutoring to many more students than could have received it individually.