Richard L. Allington's article in this issue of Educational Leadership (p. 22) inaccurately critiques previous research on the prevention and remediation of reading failure and also overlooks recent findings. The five studies reviewed by Fletcher and Lyon (1998) indicate that appropriate classroom instruction combined with intensive interventions can bring the early reading skills of 90–95 percent of the classroom population to within the average range.
This review described findings from a Title I classroom study (Foorman, Francis, Fletcher, Schatschneider, & Mehta, 1998) showing that effective classroom instruction alone can reduce reading failure to approximately 6 percent. Other recent studies show that a combination of effective classroom instruction and targeted small-group instruction can reduce the proportion of students (with an IQ of 72–122) performing below the 30th percentile in early reading skills to below 2 percent (Denton & Mathes, 2003; Foorman, 2003; Torgesen, 2002, 2004).
With respect to cost factors, current evidence shows that small-group instruction can be as effective as one-to-one tutoring for preventive instruction (Elbaum, Vaughn, Hughes, & Moody, 1999; National Reading Panel, 2000). Recent reports of significant whole-school reductions in reading failure (from 31.8 percent to 3.7 percent) (for example, King & Torgesen, 2003) also demonstrate what is possible through careful allocation of existing resources to support high-quality schoolwide reading programs. Newer classroom studies underscore the importance and cost-effectiveness of allocating resources to improve teacher knowledge. For example, recent studies of 4,872 children in high-poverty schools found reading improvement to be significantly related to professional development within a coaching/mentoring model in the classroom, above and beyond the reading method used (Foorman et al., 2003; Foorman & Moats, in press; Moats & Foorman, 2003).
We disagree with Allington's definition of the “average range” as the 45th percentile or above. We take the position that children reading above the 25th to 30th percentile (above a standard score of 90 to 93) are performing in the average range as it is traditionally defined (Sattler, 2001).
Evidence from many successful schools and from multiple research studies shows that a multitiered approach involving high-quality classroom instruction alone and in combination with targeted, small-group interventions can substantially reduce the proportion of students who struggle to read. We are making significant strides toward our goal of ensuring that each child learns to read. To support this goal, the National Institute of Child Health and Human Development funding for reading research in 2003 exceeded $52 million, a 165 percent increase since 1998.
