Repeated Reading is Part of the Science of Reading 

With all the attention that gets put on phonics vs whole-language instruction, I think sometimes other forms of evidence-based instruction get thrown by the wayside. I can think of no better example than Repeated Reading. Despite having robust scientific evidence, I almost never see anyone within the SOR community discussing the fluency instruction methodology. In general, fluency instruction is meant to help students read more accurately/quickly and is provided, once students know their letter sound correspondences, but cannot yet read fluently. Repeated Reading specifically, works by having students read the same text, (typically a paragraph or page), over and over again, as fast as they can, usually with an adult providing corrective feedback. 

 

This all being said, I get it, Repeated Reading is a tough sales pitch. When I first heard of Repeated Reading and fluency instruction, I was skeptical. I had never tried it in school, heard about it in teachers college, and quite frankly the idea of reading the same text over and over again with students sounded boring as can be. The whole idea reminds me of the 1800 school houses, in which students would recite the same readers over and over again. 

 

In 2020, I interviewed the esteemed Dr. Shanahan on the topic and he very politely informed me that I was wrong. And now having looked at the evidence, on multiple occasions, I have to admit, he was right. The research evidence for Repeated Reading is ironclad. We have 3 meta-analyses specifically on the topic and all three show results that are on average higher than the mean result in meta-analyses for phonics instruction. Moreover, the NRP meta-analysis also looked at Repeated Reading and found results similar to phonics. Take a look at the below graph, which charts these results and keep in mind that the mean result I found in my meta-analysis for phonics was .42, that the mean results for the NRP phonics meta-analysis was .45, and that the mean result for all phonics meta-analyses is .55. 

As you can see from the above results, there appears to be very strong evidence for Repeated Reading, within the literature, possibly even stronger than the evidence for phonics. That being said, while the research results for Repeated Reading tend to be higher, we have more total research on phonics, which might be skewing the picture. Whenever I bring up the research for Repeated Reading, my most intelligent colleagues always ask “But what about the assessments?” Afterall, if we have students read the same text over and over again, and then assess them on the same text, of course they will improve. However, what really matters is if those improvements would transfer to another text. 


I decided to dig a little deeper into the research. I chose to more closely examine the results of the 2017 Yoon  meta-analysis, and the 2004 Therrien meta-analysis, as these two meta-analyses had stricter inclusion standards. Moreover, the Therrien meta-analysis had the highest number of studies included, and the Yoon meta-analysis was the most recently conducted.

 

Therrien Results: 
 

Therrien Discussion:

 

Looking at these results a couple of things jump out to me. Firstly, the results for different texts are clearly lower than for the same text. However, the results for different texts are still strongly positive and higher than the mean result found for phonics, within the literature. That being said the studies that used the same text are obviously conflating the results and inflationary. It is also interesting that the results for students with learning disabilities are clearly higher than the results for students without learning disabilities. I think this suggests that Repeated Reading is likely more important for struggling and dyslexic readers than non-struggling students. It is important to note that feedback did not seem to have any meaningful impact on the results, which leads me to believe that repeated reading can even be done with very little input from teachers. Lastly, the results for coupling mastery instruction with repeated reading are obviously extremely impressive, making me believe such a practice should be a recommended one. 

 

Yoon Results:

Yoon Discussion:

I think there are several interesting observations that can be made about these results. While results for interventions in which the assessment used the same text as the intervention (no passage transfer) were much higher, the results in which the assessment was different (passage transfer) were still very significant. It is also interesting to note that there was an extremely high impact for having students read the same text, more than four times. There was also a very significant impact for reading the text to students first. Lastly, it is interesting to note that while Repeated Reading worked better in elementary school, it was still a high impact in secondary school, the same of which cannot be said for phonics instruction.

 

Conclusion:

Overall in the Repeated Reading research we see strong evidence of efficacy (regardless of assessment conditions, or student age. We see strong evidence for higher numbers of repetitions, strong evidence for instructor pre-reads, and strong evidence for the importance of Repeated Reading for dyslexic students. Critics of Repeated Reading often suggest we use less repetitive strategies to teach fluency, this seems logical, as it would definitely appear to be less draconian; however, meta-analyses of non-repetitive fluency interventions have shown much lower results.  Varied Reading is a promising alternative created by the Iowa Research center. However, there are too few studies (so far) on the topic for a proper comparison. As it stands, Repeated Reading is clearly the most evidence-based fluency intervention; moreover, the statistical evidence for Repeated Reading is actually higher than the statistical evidence for phonics and it therefore deserves its place, within the banner of the Science of Reading. That being said, we need more meta-analysis research on Repeated Reading that breaks down outcomes, according to age, and intervention duration, so we can learn how long the intervention should be used for, and in what grades. 

Written by Nathaniel Hansford

Last Edited: 2022-05-20

References: 

Therrien, William. (2004). Fluency and Comprehension Gains as a Result of Repeated Reading: A Meta-Analysis. Remedial and Special Education - REM SPEC EDUC. 25. 252-261. 10.1177/07419325040250040801. 

 

Lee J, Yoon SY. The Effects of Repeated Reading on Reading Fluency for Students With Reading Disabilities: A Meta-Analysis. Journal of Learning Disabilities. 2017;50(2):213-224. doi:10.1177/0022219415605194

 

NRP. (2001). Teaching Children to Read: An Evidence Based Assessment of the Scientific Literature on Reading Instruction. United States Government. Retrieved from <https://www.nichd.nih.gov/sites/default/files/publications/pubs/nrp/Documents/report.pdf>. 

 

Zimmermann, Leah & Reed, Deborah & Aloe, Ariel. (2019). A Meta-Analysis of Non-Repetitive Reading Fluency Interventions for Students With Reading Difficulties. Remedial and Special Education. 42. 074193251985505. 10.1177/0741932519855058.