At-risk and special education students in the 7th and 8th grades in a rural middle school in western Alabama, in the years following the introduction of the No Child Left Behind federal legislation in 2002, failed to achieve adequate yearly progress in reading. School districts are increasingly implementing flexible computer-based intervention programs to improve their students' reading achievement. Using a between-group design, the purpose of this study was to determine whether NovaNET, a newly adopted reading intervention program, enhanced the reading attainment of at-risk and special education students. Guided by constructivist theory, archived reading achievement data from the 2009-2013 Alabama Reading and Mathematics Test were analyzed for 3 consecutive cohorts of special education and at-risk students who did (n = 76) or did not (n = 73) participate in the NovaNET program. With dependent variables of reading achievement at the end of 7th and 8th grade, with independent variables of experimental-control group, gender, and general-special education status, and a covariate of reading achievement at the end of 6th grade, a multivariate analysis of covariance indicated a significant main effect associated with participation in the program (F = 4.13, df = 2, p < .02), whereas significant higher-order interaction effects pointed to differential program benefits for specific subgroups of students. Although overall effect sizes were small to modest, the results indicated that NovaNET can increase educational attainment for at-risk and special education students who are struggling with reading. This study may contribute to positive social change by providing educators with scientific data about a flexible, technology-enhanced program to promote reading instruction and achievement for at-risk general education and special-education students entering middle school.