Using RTI Data to Inform SLD Determination – Part 2 of a 5 Part Blog Series



Recent Comments


    The changes to specific learning disability (SLD) determination contained within the 2004 reauthorization of the Individuals with Disabilities Education Act expressed much of the ambiguity surrounding the field at that time. These changes came about because of the recognition that the discrepancy approach was inadequate as an identification method. At the time, the leading front-runner to discrepancy was Response to Intervention (RTI), which claimed the advantage of allowing for early intervention, for being more systematic in the attempt to ensure that learning difficulties were not a result of lack of appropriate instruction, and for relying on multiple measures that captured student growth rather than static points in time.

    RTI was not without its critics, however. Although the model did address many of the problems the field faced as a result of the discrepancy approach, it had its shortcomings. First and foremost, the research on RTI as an LD determination model was, and remains, slim. Most of the research and tools were centered on early reading, leaving significant gaps in the knowledge base about what to do for kids with learning disabilities in math or writing, as well as for kids who were late emergent reading disabled. Second, although the progress-monitoring component of the RTI model allowed for repeat assessments of student performance on a particular task, at the end of the day, SLD under this identification scheme is still defined through one measure—a lack of response. The reliance on one measure (no matter what it is) fails to fully capture the complexity of the SLD construct. Finally, whereas RTI models may provide the data that a student is not performing well and has failed to respond to instruction that is generally effective, it fails to provide an explanation for the student’s learning difficulties. This is problematic for two primary reasons: 1) it doesn’t tell us what to try next, and 2) it provides no explanation to the student and family about the nature of the underlying learning difficulty. Without such explanations, it can be difficult to design and deliver effective interventions.

    The 2004 reauthorization captured this debate around SLD determination. States were authorized to approach LD determination in a few ways. They could not require a discrepancy approach, they could use a measure of response to intervention, they could use a pattern of strength and weakness approach, or they could use some “other, research validated approach.” The federal regulations were frustrating because they left states in a quandary as to how to move forward with LD determination. Many states patterned their state policy directly on the federal language and allowed for multiple approaches. But the problems inherent in such a system are clear—the lack of a consistent identification approach calls into question the entire construct of LD, and raises important concerns about designing and delivering interventions. There had to be a better way.

    In Idaho, a small team of stakeholders (SLD team) convened to tackle the issue. Drawing on the existing research and literature base, the team decided to start at the beginning, which meant analyzing the current definition of learning disability. What they found was that the federal definition had remained surprisingly consistent for decades, and that it contained a few, salient characteristics. First, a learning disorder was defined as a disorder in one of the basic psychological processes. Second, the disorder manifested itself in an imperfect ability to learn in one of several academic areas. Finally, the learning difficulty was not due to other, competing factors such as vision or hearing impairments, economic disadvantage, or cultural differences.

    The SLD team decided to align its operational definition of LD to the salient characteristics of the federal definition. This meant that a multidisciplinary team (MDT) considering whether a child had an LD would need to investigate and provide evidence for the three salient characteristics. The first marker would likely be the student’s low achievement in one or more academic areas of concern—this would provide evidence that the student was experiencing an imperfect ability to learn. RTI data collected through benchmarks, progress monitoring, and student outcomes provided an excellent fit for documenting the child’s learning difficulties. Individual student-level data can establish that the child is not achieving to grade-level standards and is not responding to intervention, whereas class-, grade-, and school-level data can help establish that the instructional program is generally effective for most students.

    The exclusionary criteria that had long been a required component of eligibility decisions would be considered as relevant for each child, and documentation would need to be provided to help rule out competing explanations for the child’s low achievement. In addition to screening for vision and hearing issues, RTI data would help the MDT make a decision about the extent to which other factors played a part in the student’s learning difficulties. For example, if the student came from a low socioeconomic status household, then the RTI data examining the performance of other students from similar backgrounds could be reviewed to determine whether the learning environment appeared to be supportive of student learning needs. If students from similar backgrounds were making progress in response to instruction, but the student of concern was not, it would give the MDT more confidence that the learning difficulty was not due to an external factor.

    Now the SLD team needed to address the issue of determining whether there was a basic disorder in a psychological process. Although this had long been a part of the definition, it had never been a routine part of eligibility determination. The SLD team reviewed the research and conducted a meta-analysis on the existing research examining the relationship of cognitive processing deficits and academic achievement. The research suggested that there was credible evidence that students with learning disabilities tend to have cognitive processing deficits in one or more areas that underlie academic performance; however, clear connections to establish which processing areas underlie which academic area and the size of the deficit are less well determined.

    Based on these three characteristics, the SLD team developed a state LD eligibility policy that would require MDTs to provide evidence to support each of the three areas. A full version of the state policy can be found in theIdaho Training Clearinghouse, but in general, an MDT must document that the child is experiencing low achievement in one or more academic areas of concern and has failed to respond to an evidence-based intervention. The MDT must also document the performance of the student’s peers. This helps ensure that instruction is being considered first. The child’s low achievement must be confirmed by a standardized academic achievement measure that is related to the specific area of academic concern (e.g. basic reading, reading comprehension, written expression). Then the team must determine the underlying cause for the student’s learning disability through an assessment of cognitive processing areas suspected to underlie the student’s specific difficulty.

    What is unique about Idaho’s approach to LD identification is that it captures all three of the salient characteristics of the federal definition and relies on a clinical approach rather than on hard and fast, predetermined cut scores. The clinical approach was adopted in favor of specifying cut scores because there is insufficient evidence to support hard and fast cut scores, and because learning disabilities are heterogeneous and exist along a continuum.

    The clinical approach in theory is better aligned with the complex, heterogeneous, and individualized nature of LD. In practice, of course, this approach presents implementation concerns that need to be addressed to ensure reliable/consistent application of the policy.
    In the next blog post, I’ll describe the system of support that the SLD team developed statewide.
    Back To Top