Melanie R. Kuhn and Steven A. Stahl's article, Fluency: A Review of Developmental and Remedial Practices, from The Journal of Educational Psychology , 2003, 95(1), is an often quoted reference in the fluency literature. After describing Chall and Ehri's proposals on stages of reading development, they look at research in the field. They broke the research into three main groups: repeated reading, assisted reading and classroom interventions. The lack of control groups prevented the use of meta-analysis of the research so the authors used a more simple counting strategy.
Ultimately, the authors concluded that a focus on fluency improved reading as assessed by comprehension. They found that the majority of the interventions for remedial students did not improve reading at a rate faster than their reading proficient peers. This was a concern since we would like to identify methods of intervention that improve reading rates faster than for average peers. The challenge is that one definition of a learning disability is that it takes longer to learn the material than typical peers. Consequently, without additional time in reading, it may be impossible to bring disabled students to grade -level performances.
The chart of studies included in the survey is a valuable tool. Interestingly, it reveals the dearth of research regarding older students. Only three of the studies included involved students at the secondary or college level and only one of them involved students with reading disabilities. This means that the conclusions of the review may not be applicable to older students.
Hollingsworth conducted two studies that had intriguing results. One showed no improvement in comprehension when fourth grade at level peers were given fluency instruction. The other study examined remedial fourth and sixth grade students. They showed above average growth in reading comprehension using assisted reading.
According to the results of the review, fluency instruction proved most valuable to students reading between a late preprimer level and a late second grade level. The authors said "It is not the repetition that leads to the effect [improved reading skills] but the amount of time spent reading connected text." (p. 17) This falls into line with the idea that time equals progress more than a specific strategy. They also concluded that the level of reading that was most successful with progress was the individual student's instructional reading level.
Research that has been conducted in the decade since this review seems to indicate that intensive remedial programs with fluency components can be effective at improving reading skills in secondary students as well as with elementary students. Meta-analysis of current research would be interesting to evaluate.