Melanie R. Kuhn, Paula J. Schwanenflugel, Robin D. Morris, Lesley Mandel Morrow, Deborah Gee Woo, Elizabeth B. Meisinger, Rose A. Sevcik, Barbara A. Bradley and Steven A. Stahl's manuscript, Teaching Children to Become Fluent and Automatic Readers, published in the Journal of Literary Research 2006, 38(1), describes a study in which the authors attempted to tease out if wide reading or repeated reading is a better reading instructional model than other traditional methods. They chose second grade classrooms in New Jersey and Georgia in which to perform their year long research.
Overall both wide reading and repeated reading models proved superior for developing word automaticity and comprehension over more traditional approaches. The authors point out that gains for wide reading occur earlier than those for a FORI (a specific text repetition method- Fluency Oriented Reading Instruction) but that over the course of the year the gains level out to be equitable for both. One of the things they pointed out was that since most texts appropriate for second grade readers have a significant proportion of sight words and word repetition, narrowing in on material appropriate for that grade level resulted in significant exposure to very similar word lists.
In their study they included all students in the classes. For the lowest performing six students in each classroom, reading interventions focused on foundational sight word and decoding skills were provided for 45 minutes per day in addition to full participation in the class reading lessons. This additional scaffolding enabled these struggling learners to interact in the whole class lessons and activities and provided remediation as needed. Further, both the wide reading and FORI methods presented the teacher reading the material the first time through to provide support for reading the grade level texts. The authors identified the appropriate level of difficulty to be on materials that the students could read 85% of the words correctly. When students struggled with material, additional practice time was provided by sending the reading home for practice.
The major difference between the control and the experimental groups was the amount of time spent reading contextual material. Students in the experimental groups spent over two hours a day reading. It is instinctively obvious that that more time spent reading the greater the reading progress.
I would caution the reader of the manuscript- it is not in final published form. There are a significant number of typos including comma/period exchanges and missing spaces. They also quote the amount of time reading per day in the experimental group as 2040 minutes which equals 34 hours, clearly there is a confusion here. This increases the challenge of reading; digging out the final published form may be worth the effort.