In this article, I wanted to continue our series on whether or not 95% of students could learn how to read. The figure 95% of students can learn how to read originates from two literature reviews. The first one was published by Patricia Mathes and Carolyn Denton in 2002 and the second by Joseph Torgesen in 2009. Both literature reviews seem to suggest that more than 95% of students could learn how to read (usually defined by reaching the 30th percentile on the Woodcock Reading assessment), provided they were provided with 80 hours plus of small group, tier 3 instruction. Last week, I re-analyzed their results and specifically controlled for the impact of systematic phonics tier 3 instruction, within the studies covered by Torgersen and Mathes. I found that if struggling readers were given 80 hours of tier 3 systematic phonics instruction, 97.5% of all students within the overall class could learn how to read on grade level.

However, there is a catch to all three literature reviews. They were all based on the idea that only 18% of readers in the school started below the 30th percentile. However, the NAEP basic marker is roughly the 25th percentile and currently 37% of students in the USA are reading below this benchmark. This means that in order for it to be a realistic goal for policymakers to suggest that 95% of students can learn to read, we would need to show that up to 82% of readers can get to grade level based on core instruction alone. Unfortunately, there is no evidence available to answer this question because the vast majority of reading studies use mean difference and standard deviation, to measure results not the percentage of students on grade level. 

Why don't we see academic research using "grade level benchmarks'' as a metric?  

It’s a poor scientific metric because it uses an arbitrary cut off mark that is inconsistent across populations, tests, and contexts.It does not account for students who are far above grade level, nor does it account for if students are only slightly above grade level. Conversely, an effect size is primarily used to measure the magnitude of effect that a treatment provides to one group of students over another. For example, we could have 100% of students in one class, on grade level and 0% in another, but the mean difference could still be only 1%. The outcome would look significant; however, in reality, this difference could be entirely statistically insignificant. That said, the percentage of students reading on grade level is more commonly used in non-academic circles. It’s easier for both educators and politicians to understand. Consequently, policies are often based on metrics related to such measurements. 

This disparity means, that despite the fact that the percentage of students reading on grade level is likely a widely discussed metric, amongst policymakers and educators it is difficult to find a meta-analysis or even a rigorous analysis of the subject. The literature reviews currently on this topic, were all based on small-scale, tier 3 intervention studies. I wanted to answer the question, of what percentage of students could learn to read on grade level, from just core instruction alone? 

To answer this question, I went through the database of studies that we have gathered for Pedagogy Non Grata. First I eliminated all studies that were not based on purely core instruction. Then I looked through the studies that showed effect sizes of .40 or higher (I chose this cutoff point, as I was hoping to measure the potential result and not the average result. In total, 28 studies that met this inclusion criteria. However, only 11 studies included the percentage of students at grade level. Two of these studies were RCTs, four of these studies were quasi-experimental, and five of these studies did not use control groups. After this I found two additional studies, one RCT was sent to me, by Dr. Rachel Schechter, who has been collaborating with me on this project. The final study was included in the 2009 literature review (Torgesen 1999), in which the control group received systematic phonics as part of their core instruction. In total, I found 13 core instruction studies, in which the percentage of students at grade level was measured. Three of these studies looked at systematic phonics, and ten of these studies looked at just providing additional supplementary phonics instruction to business-as-usual instruction. I would have liked to also look at balanced literacy studies on this topic. However, I was unable to find any, as most balanced literacy studies only focus on tier 3 instruction.

On average 82.92% of students reached grade level in these studies. However, I have broken down these results more in the following graph.