The Windward School's Head Lines Blog

Head Lines appears in print in The Beacon. To view the full, most recent edition of The Beacon, please visit here.

Response to Intervention: Let’s Get It Right
By John J. Russell, EdD

It has been almost 15 years since Congress passed legislation authorizing the use of Response to Intervention (RTI) as a means for identifying and remediating students with language-based learning disabilities. There are four common elements of RTI: high quality, research-based instruction in general education; continuous progress monitoring; screening for academic and behavior problems; and multiple tiers of progressively more intense instruction (Office of Special Education Programs, 2006).

These four elements of RTI are typically delivered to students through three tiers of intervention. Tier 1 calls for qualified teachers to provide a research-based reading program to all students in general education classrooms. Students who do not make appropriate progress at the Tier 1 level are moved to Tier 2 where they receive special education services in their schools. Students who do not respond sufficiently to Tier 2 interventions are then eligible for Tier 3 and placement in special education schools. 

Since its inception, this very promising tool has been used to varying degrees of success. A review of the development of RTI and the evolution of its implementation provides important insights into its benefits and shortcomings.  

A Brief History: The Genesis of RTI


The Education of Handicapped Children Act was passed in 1975 and amended in 1990, becoming the Individuals with Disabilities Education Act (IDEA). While both of these laws resulted in significant improvements for students with disabilities, they also contained a major flaw by requiring an IQ-achievement discrepancy to identify students with language-based learning disabilities (Fuchs & Fuchs, 1998). The discrepancy model compares assessments of a child’s intellectual ability with the progress the student is making in school.

As a result of IDEA, unexpected underachievement based on IQ became the primary criterion for identifying students with learning disabilities. Unfortunately, this methodology is at best defective and at worst harmful to students. The reality is that IQ and academic performance are not well correlated. In fact, the correlation between measures of intellectual ability and academic achievement rarely exceed .60 (Sattler, 2001). Almost immediately, it became very clear that establishing a discrepancy between intelligence and achievement is not sufficient for assessment or intervention purposes, especially for students in the early grades (Restori, Katz and Lee, 2009).

Despite the mounting evidence of the shortcomings of the IQ-achievement discrepancy model, the reauthorization of IDEA, which was entitled the Individuals with Disabilities Education Improvement Act (IDEIA, 2004), allowed states and school districts to choose between using the traditional IQ-achievement discrepancy model or Response to Intervention (RTI) for identifying students at-risk for a Specific Learning Disability (SLD). It is important to note that the 2004 amendments established only that local education agencies “may” use RTI in determining whether a child qualifies for special education services by being identified as having SLD. While each state has the option to either require or allow RTI for SLD identification, a state may not prohibit it.

According to, 39 states still allow their school districts to use the discrepancy model while 11 states forbid it. For example, it was not until 2012 that the New York State Department of Education banned the use of IQ-achievement discrepancy for identification and required the use of RTI. The Department’s website states: "Effective on and after July 1, 2012, a school district must have an RTI process in place as it may no longer use the severe discrepancy between achievement and intellectual ability to determine that a student in kindergarten through grade four has a learning disability in the area of reading."


The Response to Intervention Model


RTI has garnered support from many significant sources. The framework of RTI is very similar to the recommendations that the National Reading Panel made in 2000. In his meta-analysis of the effectiveness of educational strategies, the Australian researcher, John Hattie (2016) found Response to Intervention to be one of the most powerful practices with an effect size of d=1.07. Other researchers (Swanson, Tran, Sanchez & Arellano, 2011) have found RTI effect sizes ranging from .45 to 1.53. An effect size of 0.8 or higher is considered large (Cohen, 1988). With these values, RTI should be a potent intervention. In his book Language at the Speed of Light (2017), Mark Seidenberg states, “RTI is a thoughtful, logical, well-designed program.” By these measures and other very positive reports, Response to Intervention should be producing significant improvement in reading for both the general education and special education populations, but it is not.


Reading Performance Since RTI

The acquisition of reading skills by students is regularly assessed at the international, national, and state levels. In the 15 years since the introduction of RTI, an analysis of these measures paints a bleak picture.

The Programme for International Student Assessment (PISA) is a triennial international survey that evaluates education systems worldwide by testing the skills and knowledge of 15-year-old general education students. Since the inception of PISA in 2000, the scores of United States students on the PISA Reading Literacy portion of the test have resulted in a steady decline in the ranking of the United States compared to other members of the Organization for Economic Cooperation and Development (OECD). In 2000, the United States ranked 7th; in 2009, it ranked 17th; and in 2017, it ranked 24th.

The National Assessment of Education Progress (NAEP), also known as “the nation’s report card”, is the largest nationally representative and continuing assessment of America’s students’ achievement in a range of subject areas. Since 2004 when RTI was instituted, the performance of general education students has improved somewhat but remains consistently subpar. The performance of students with disabilities has also improved slightly, but can only be described as abysmal. It should be noted that the classification “Students with Disabilities” includes all disabilities; however, 70 to 80% of that group are students with language-based learning disabilities—most notably dyslexia.

With 60% of general education students and a stunning 88% of learning disabled students failing to meet proficient reading standards, it is entirely reasonable to ask how an evidenced-based program with a significant body of research confirming its efficacy would produce no better results than have been observed after 15 years of practice. Researchers offer a number of insights into the reasons for the failure of RTI to meet expectations for general education students and students with disabilities. Two factors stand out for the negative impact they have on teaching students to read: the use of reading programs and instructional methodologies that are not research-based and inadequately trained teachers.


Lack of Clarity in the Definition of “Research-Based”

The movement toward research-based practices was partly a response to the pseudoscientific and unproven interventions that have plagued special education for years (Kozloff, 2005). Unfortunately, the term “research-based” (also known as “evidence-based”) is not definitively defined in the RTI legislation, so states, school districts, and individual teachers are left to decide what is research-based and what is not. Mistaken beliefs about teaching and learning often originate from schools of education and from education officials at the state and local levels who, despite decades of accumulated evidence, continue to incorrectly believe that systematic and explicit instruction is harmful to learning, that an eclectic approach is best, and that teachers should be innovative (Seidenberg, 2017).


Publishers of textbooks and reading programs have seized upon these misguided beliefs by applying the label “research-based” arbitrarily. For example, publishers have, to a large degree, abandoned the use of the widely discredited approach called “Whole Language” and have instead adopted “Balanced Literacy” touting these reading programs as research-based when in fact they are not.

As early as 2000, Louisa Moats chronicled this deception in her seminal piece, Whole Language Lives On: The Illusion of “Balanced” Reading Instruction, in which she cautions, "It is too easy for practitioners, while endorsing “balance,” to continue teaching the whole language without ever understanding the most important research findings about reading or incorporating those findings into their classroom practice. Wrong-headed ideas about reading continue to characterize textbooks, reading course syllabi, classroom instructional materials, state language-arts standards, and policy documents."

In addition to balanced literacy instruction, which is simply a repackaging of the whole-language, many other popular education interventions are unsupported by evidence. In fact, the use of unproven practices proliferates in schools across the country (Miller & Sawka-Miller, 2010). The widespread use of interventions that are not supported by research partially explains the underperformance of RTI in improving literacy. Jason Travers (2016) described the consequences of these unsupported practices, which are often publicized as “innovative,” stating, "Although many unproven or pseudoscientific interventions might appear relatively benign at first glance, it could be argued that every ineffective intervention is associated with some degree of harm. The main problem is that a tried intervention is only revealed to be a failure after the investment is made; instructional time is permanently lost and educational benefit is not conferred. Implementation of an intervention that failed to confer benefit means resources were wasted and a student’s opportunity to learn (i.e., time) has been permanently lost. Every student with a disability has a finite amount of time to receive special education services, and professionals are ethically obligated to maximize the impact of these limited services. It may seem that a few weeks of time exploring whether an intervention works has only minimal harm, but a small amount of time lost to ineffective instruction can accumulate over time to a significant loss of potential educational benefit."


Inadequately Trained Teachers

“Parents who proudly bring their children to school on the first day of kindergarten are making a big mistake. They assume that their child’s teacher has been taught how to teach reading. They haven’t.” (Seidenberg, 2017).

Many Americans think that the ability to teach is more the result of innate talent than training, but recent research clearly demonstrates that the best teachers are made, not born (Goldhaber, Liddle, & Theobald, 2013; Michelli, Dada, Eldridge, Tamim, & Karp, 2016). Tragically, teachers are not being adequately prepared in colleges and universities to become the effective teachers required for RTI to reach its full potential.

There remains a significant disconnect between the preparation teachers need to be successful in implementing RTI and the preparation they actually receive in their pre-service and graduate education courses (Walsh, Glaser, & Denne-Wilcox, 2006). In 2017, the National Council on Teacher Quality (NCTQ) released its annual review of teacher preparation programs in the United States. As was the case in previous studies, once again colleges and universities were cited for their substandard preparation of teachers. Of great significance to all parents, and especially to parents of learning disabled students, the NCTQ evaluated 717 college and university pre-service teacher education programs that prepare teachers to teach students who are identified as “struggling readers” and found that 75% of these programs did not meet the basic standards set by NCTQ. Although lack of adequate teacher preparation disproportionally harms students with disabilities, it has serious negative consequences for all students as well as for the teachers themselves who must suffer the professional and emotional burden of not being able to adequately support the children they work so hard to teach.

A recent study, conducted in 2015 by John Hattie of the University of Melbourne confirmed the dominant effect that teacher quality has on student performance in general and by extension on initiatives like RTI. He conducted a meta-analysis of more than 65,000 research papers on the effectiveness of hundreds of interventions on the learning of millions of students and found that what matters most is teacher expertise. The most powerful ways to improve student learning identified by this meta-analysis all depended on what teachers did in the classroom. Every day teachers make hundreds of instructional decisions based on what they have been taught in their pre-service and graduate school programs. Regrettably, few teachers are exposed to research that is relevant to their jobs and most are ill-prepared to critically assess scientific claims, leaving them vulnerable to fads and fallacies in their instructional decision making as they search for a program or methodology that will help them teach their students to read and write (Siedenberg, 2012).



If RTI is to deliver on its promise to improve reading achievement for general education students, to efficiently identify students who need special education services and to provide effective interventions for special education students, several steps must be taken.

First, all teaching materials and teaching practices that are labeled “research-based” must be held to certain standards. Citing Skinner (1953), Travers (2016) offers the following criteria necessary to warrant the label “researched-based” or “evidence-based.” "Evidence-based special education depends on the acquisition of robust empirical findings obtained via meticulous experimentation. The processes of empirical inquiry are necessarily accompanied by a set of attitudes that emphasizes valuing facts over authority, accepting evidence regardless of conflict with strongly held beliefs, and abstaining from acceptance of a claim until compelling evidence is available (Skinner, 1953)."

Second, the quality of teaching must be improved. Schools need to recognize that during their undergraduate education teachers have, in most instances, not received the foundational knowledge necessary to teach reading. To improve the quality of teaching reading, comprehensive professional development is an absolute necessity. Dedicated, conscientious teachers can mitigate deficiencies in their preparation through professional development, but only if professional development programs are more rigorous and of a better quality than the undergraduate and graduate programs that are responsible for the deficits in the first place. In response to this reality, The Windward School established a professional development program that is comprehensive, demanding, and extremely effective in closing the knowledge gap between research and teaching practices. Other schools have followed a similar path in their efforts to improve teacher quality.

For RTI to be truly effective, highly qualified, trained teachers must use instructional practices and programs that have been rigorously validated as evidence-based.


  • teacher training