A quasi-experimental pretest/posttest group design was used to determine whether computer-assisted fluency instruction is as effective as print-based, teacher-led fluency instruction in improving fluency, vocabulary, and comprehension skills in third grade students experiencing delayed fluency development. Fifty participants were randomly assigned to 1 of 3 conditions: a teacher-led group, a computer-assisted text-equivalent group, or a computer-assisted time-equivalent group. The same fluency program was used for all three groups. An analysis of covariance (ANCOVA) revealed no significant differences across groups on 2 of the 3 outcome measures (i.e., fluency and comprehension), but significant differences between the text equivalent and time equivalent treatment groups occurred in the area of vocabulary. Limitations and recommendations for research and practice are discussed.