Responsiveness-to-Intervention Symposium

December 4-5, 2003 * Kansas City, Missouri

The National Research Center on Learning Disabilities sponsored this two-day symposium focusing on responsiveness-to-intervention (RTI) issues. The speakers, discussants, and participants assembled represented the wide diversity of individuals with a vested interest in LD determination issues. Advocates, instructional staff, researchers, and state-level education officials brought their collective and considerable expertise to the discussions.

Executive Summary of the NRCLD Symposium on Responsiveness to Intervention

Download Executive Summary (pdf)

The National Research Center on Learning Disabilities is charged with conducting research on the identification of learning disabilities; formulating implementation recommendations; disseminating findings; and providing technical assistance to national, state, and local constituencies. We have an obligation to raise the awareness of policymakers, parents, special education administrators, and diagnosticians in understanding research findings and using that information when making decisions. We believe that the best education policy emerges when leaders consistently seek and use quality data, interpreted in light of the attitudes, values, and beliefs of interested parties, to drive their decisions.

In 2000, the U.S. Department of Education's Office of Special Education Programs (OSEP) began planning a series of activities, collectively called the Learning Disabilities Initiative. The Learning Disabilities Initiative was designed to establish a credible foundation of research related to learning disabilities identification upon which future decisions could be based. The collective goal of these activities is to identify and implement improved procedures for LD identification. In its systematic approach to accomplishing this goal, OSEP has taken the following steps:

Commissioned a set of papers addressing critical issues related to LD determination.
Convened the Learning Disabilities Summit: Building a Foundation for the Future in August 2001.
Conducted several roundtable discussions with key organizations and stakeholders, including researchers and members of the National Joint Committee on Learning Disabilities (NJCLD).
Established the National Research Center on Learning Disabilities (NRCLD) to conduct research in this area, make recommendations, and provide the assistance required to help schools improve the manner in which they assess and identify students with learning disabilities.

Many concerns have been raised about current procedures for assessing and identifying students who have learning disabilities, including the following: the exponential increases in the number of students who are considered to have learning disabilities, the reliance on IQ tests, the exclusion of environmental factors, the inconsistency in procedures and criteria within school districts and across states, and the reliance on aptitude-achievement discrepancy formulas and the manner in which they are used. These concerns prompted OSEP's LD Initiative.

Many leading researchers have proposed and supported responsiveness to intervention (RTI) models as a possible alternative assessment method for identifying students with LD. These researchers believe RTI addresses many of the concerns with current practices, including the need to reduce the number of students misdiagnosed as having LD and to improve educational outcomes for these students. RTI consists of a multi-tiered prevention model with the primary intervention consisting of a general education program based on evidence-based practices, secondary interventions involving intensive, short-term interventions, and tertiary interventions in the form of specially designed instruction or special education.

Although RTI models appear to hold promise, many unanswered questions surround the application of RTI on a broad scale. In December 2003, NRCLD and OSEP sponsored an invitation-only symposium to begin to address these issues. NRCLD selected six questions to address in this first symposium. The questions were selected on the basis of their relevance to the topic, their timeliness, and their potential to identify areas of concern that would need to be addressed before considering the adoption of RTI as an LD identification method.

In selecting potential speakers to address these questions, NRCLD looked for researchers who had significant experience in RTI and who were recognized as leaders in the field. In addition, NRCLD sought speakers who would bring a diversity of opinion and balance to discussions during the symposium's two days.

Equally important to these discussions were the perspectives of the invited participants. NRCLD invited representatives of the research community, parents, practitioners, administrators, and diagnosticians. Leaders from each of the 11 NJCLD organizations were invited.

This document provides a broad overview of the topics discussed during the symposium's six sessions. Detailed information--including papers, PowerPoint presentations, transcripts, videos, and other supplemental materials--is available on NRCLD's web site, www.nrcld.org.

Session One: How should screening for secondary intervention occur?

One of the main criticisms of current approaches to LD identification is that they are "wait to fail" models, in that discrepancies between achievement and aptitude usually do not appear until a student is in third grade or later, long past the stage for receiving early interventions. RTI proposes to address this issue by implementing a multi-tier process in which students are screened for at-risk indicators for reading problems and thereby are eligible for receiving secondary interventions. Screening is the mechanism for identifying struggling readers who lack prerequisite skills or who acquire reading skills at a pace that puts them at risk for unsatisfactory outcomes. Although general procedures for screening are clearly defined in an RTI model, several questions about the specifics of this process remain. In this session of the symposium, presenters were asked to discuss alternative screening procedures.

Presenters outlined several considerations for developing a successful screening procedure for secondary intervention. Paramount to any system is the requirement to minimize the number of students who mistakenly are not identified as in need of early intervention. Conversely, another issue raised during this session is that although early identification is the desired goal, the main problems with early screening include a tendency to mistakenly identify too many children as at risk, in part due to the failure of screening instruments to account for student growth. If screening is implemented as early as kindergarten, current research suggests that as many as 50 percent of students may be over-identified as being at risk. Although virtually all practitioners would advise erring on the side of over-identification, in a time of limited resources, the reach of that safety net may be too broad.

Another important concern addressed was ensuring the screening system is efficient, reliable, and integrated with the continuing assessment and instructional system already used in the school or district. Finally, presenters in this session pointed to several existing models of RTI, some of which have been implemented on a wide scale. As we continue to investigate issues of implementation in an RTI model, consideration of the concerns outlined in this session may assist educators in adopting screening mechanisms that are supported by research.

Session Two: How should secondary intervention be formulated?

Secondary intervention in an RTI model has been conceptualized in two primary ways. One includes the use of standard treatment protocol approaches that rely on the implementation of evidence-based instructional principles. The other is based on a problem-solving approach that relies on a collaborative effort between general and special educators to identify individual needs of students who require additional support. In this session, presenters were asked to discuss secondary intervention based on their first-hand experiences and research.

One presenter described how a problem-solving model is used to serve nearly a quarter of the students in Iowa. The process involves four steps:

Problem definition/analysis (What is the problem and why is it happening?)
Planning (What are we going to do about it?)
Implementation (Are we implementing as designed? Is the student making progress?)
Evaluation (Is our plan working?)

The presenter also described factors that support the quality of the model, including leadership, modifications to job descriptions and paperwork to incorporate problem-solving components, and development of a data collection procedure that is used systemwide. This model is not used to determine learning disabilities, but rather special education eligibility.

Two presenters discussed studies that focused on secondary interventions for first graders. The first study investigated the use of a dual-discrepancy approach to identifying children who were not responsive to a classwide reading program. Dual discrepancy--meaning a student's performance and growth rate are both substantially below those of average readers--seems to hold promise in distinguishing nonresponders from responsive at-risk and average-achieving children. The study also examined three increasingly individualized treatments for students identified as unresponsive. Although no statistically significant difference was found between groups in the three treatments, effect sizes suggest that one-to-one tutoring is most promising for reducing unresponsiveness.

In the second study, researchers designed interventions based on evidence that small student groups and low student-teacher ratios combined with explicit phonics-based instruction are effective. Initial results indicate significantly larger growth in the experimental group of secondary-level at-risk students as compared to controls in schools that did not employ a three-tier model or early screening. The study also identified critical features of the multi-tiered model, including leadership of the principal, involvement of the school's entire faculty, collaboration on data collection and decision making, and continuing consultation.

A discussant summarized the essential features of a three-tier model and the necessary system changes that need to occur.

Session Three: What are the feasibility and consequences of RTI?

As the nation seeks improved, research-based ways of identifying students with learning disabilities, understanding how a given identification system will be implemented on the front lines is important. NRCLD posed this question to begin to develop an understanding of such issues--in relation to RTI models--as changing roles of teachers and diagnosticians, responsibility for implementation, consistency of decision making, and consequences of altering current procedures, especially in dropping cognitive measures as part of LD assessment.

Two presenters and two discussants offered their perspectives on these questions. All agreed that the educational welfare of students is the primary concern in any discussion of these issues. They were divided, however, in their opinions regarding whether RTI models as they have been implemented to date in isolated projects are feasible on a large scale.

The main thrust of those in favor of RTI is that current practices of LD identification and delivery of services to students are ineffective and do not improve student outcomes. They see RTI's emphasis on continually monitoring student progress to help shape instruction to meet student needs as a positive alternative with the potential to improve educational performance for students with learning disabilities.

Other speakers expressed concerns about several aspects of RTI, including its narrow focus. Presenters argued that RTI models do not consider several accepted characteristics of LD and therefore will not move the field closer to understanding LD. They warned that the use of RTI may result in losing the concept of "learning disability" altogether. Other concerns raised included the potential cost of moving all schools nationwide from their current method of LD determination to an RTI model, how to ensure RTI models are used as intended for best results, and how professional development will be structured. In addition, speakers expressed concern about how the use of an RTI model will affect general education teachers, fearing that the focus on standardization in an RTI model will limit general education teachers' flexibility to respond to individual classroom needs.

Session Four: How should "unresponsiveness" to secondary intervention be operationalized in an RTI approach to LD identification?

Although secondary intervention is expected to provide early support for students identified as at risk for reading difficulties, research demonstrates that some students do not respond to evidence-based reading instruction. For these students, or "non-responders," tertiary interventions--which include specially designed instruction, or special education--are provided. Because the hope is that an RTI model will help reduce the misidentification of students with LD, an important yet complex task associated with this model involves determining how to define "unresponsiveness."

In this session, presenters reviewed data from current research studies in identification. A discussant synthesized these findings and offered recommendations about how to determine "unresponsiveness" in an RTI approach to LD identification.

Most importantly, the notion seemed to emerge that "unresponsiveness" is dependent on the context of individual educational settings. In other words, determining inadequate response to instruction is inextricably linked to addressing (a) which students are in need of additional support, (b) what the additional support should look like, (c) when the additional support should be implemented, (d) the length of time the additional support should occur, and (e) who should provide the additional support.

Two methods for determining "unresponsiveness" to secondary intervention were presented. The first method relies on a system of performance markers for early literacy skills. In this method, first-grade students are targeted for additional supports based on the number of words per minute they achieve reading grade-level materials. The second method relies on a gated system in which "responsiveness" is defined by performance in a typical range on standardized tests.

Although RTI is most commonly discussed in terms of its use at the elementary level, one presenter discussed its use with older students. Implementation with older students will depend on identification of deficits in reading comprehension, underscoring the need for addressing the present limitations associated with measuring and improving reading comprehension.

Session Five: How many tiers are needed within RTI to achieve acceptable prevention outcomes and to achieve acceptable patterns of LD identification?

RTI generally has been described as a three-tier model featuring a continuum of increasingly intensive, specialized instruction. In this session, presenters were asked to consider what interventions in Tier 3 and beyond contribute in terms of student performance and patterns of LD identification. The three presenters in this session described two controlled research studies and one large-scale project that used RTI models. Each presenter described the number of tiers used in his or her project, the approach to instruction in each tier, and outcomes over several years of implementation. Although implementation and instructional methods differed among the three projects, all presenters described a continuum of interventions matched to student needs and ultimately leading to identification of students with LD.

A discussant synthesized the findings from these projects to address two questions:

How many tiers are needed within RTI to achieve acceptable prevention outcomes?
How many tiers are needed within RTI to achieve acceptable patterns of LD identification?

How many tiers are needed within RTI to achieve acceptable prevention outcomes? Evidence presented during this session indicated students who received both Tier 2 and Tier 3 interventions showed significantly greater gains than control groups. In one study, 92 percent of the entire student population responded to Tier 1, Tier 2, or Tier 3 interventions. In reviewing these studies' results, the discussant found that 50 to 75 percent of students in Tier 2 responded to instruction. In these studies, Tier 3 represented increasingly intense and explicit instruction but did not correspond to special education.

How many tiers are needed within RTI to achieve acceptable patterns of LD identification? The answer to this second question, based on the data presented, was less clear. The discussant found little evidence to support Tier 1 alone as a sufficient means of determining special education eligibility. He found 25 to 50 percent of students in Tier 2 were "nonresponders," while in one study, seven out of ten students in Tier 3 eventually were identified for special education services. Data from the same study indicated that 6 percent of the student population in Tiers 1, 2, and 3 were identified for special education services.

Session Six: What are alternative models to LD identification other than RTI?

The primary purpose of the RTI symposium was to investigate the use of RTI as a means of LD identification. However, organizers also recognized the need to consider critiques of the RTI approach and alternate viewpoints. Although RTI addresses some of the shortcomings identified in the process undertaken 30 years ago to answer questions of how to identify LD and control its prevalence, the answers to these questions are far from settled. This final session of the symposium invited presenters to offer their perspectives on other assessment methods that could be useful for schools, other factors important for accurate LD identification, and ways to improve LD identification. Presenters also offered their critiques of current practices of LD identification and their concerns and views about RTI, alternative models, and research-to-practice issues.

Although criticism of the aptitude-achievement discrepancy method of LD identification is widespread, many special education researchers counter that what separates the notion of learning disabilities from low achievement in general is that the low achievement in students with LD is unexpected in light of the student's aptitude (IQ). Advocates of this position argue that low achievement that is not unexpected or that is due to environmental, cultural, or economic differences, although a serious concern, should not necessarily be addressed through a special education system designed for students with disabilities. Adherence to the belief that a student with a learning disability is fundamentally different from a student with low achievement has led several researchers to propose alternatives to RTI that include criteria related to the unique aspects of learning disability. Alternative models included those based on operational definitions of LD and models focusing on psychological processing deficits. Inherent in both approaches to LD identification is a belief that RTI fails to address the multifaceted nature of LD.

Conclusions

The reauthorization of the Individuals with Disabilities Education Act provides an opportunity to address concerns about the appropriate identification of students with specific learning disabilities. Presently, states and school districts employ many different methods to make a determination of whether a student has LD. This variation in methods raises concerns about equity, accuracy, timeliness, outcomes, feasibility, and consistency.

OSEP has publicly stated its commitment to addressing the shortcomings of the long-criticized approaches to LD identification. Regardless of the models chosen to supplant, revise, or complement these approaches, change at the federal level in the diagnosis of the largest category of students in special education will present many challenges to practitioners at all levels.

NRCLD's mission is to conduct research on these issues, including the use of RTI and other alternative methods of LD identification. In addition to the December 2003 symposium, NRCLD and the six federal Regional Resource Centers are engaged in a process to identify and evaluate RTI models currently in use. The project will explore the extent to which RTI is used and the feasibility of using current sites as training venues for other schools and districts in the process of adopting an RTI model.

As a result of our findings from this event and other related studies--and recognizing that policies are matters of science and values--NRCLD will develop recommendations that promote policies and practices that are aligned with current scientific knowledge. Science-based findings can improve the development of policies that further shared goals for the public as a whole and particularly for students identified as having disabilities.

Authors' note: Opinions expressed herein are those of the authors and do not necessarily reflect the position or policies of the Office of Special Education Programs or the U.S. Department of Education, and no official endorsement should be inferred.

In citing material from this brochure, please use the following:
National Research Center on Learning Disabilities. (2004). Executive summary of the NRCLD symposium on responsiveness to intervention [Brochure]. Lawrence, KS: Author.

NRCLD is a joint project of researchers at Vanderbilt University and the University of Kansas. Funding is provided by the U.S. Department of Education, Office of Special Education Programs. Project officer: Renee Bradley. Award No. H324U010004.

Executive summary prepared for NRCLD by Julie M. Tollefson, Evelyn Johnson, and Sara E. Byrd. October 2004

The symposium was made possible by the support of the U.S. Department of Education Office of Special Education Programs. Renee Bradley, Project Officer. Opinions expressed herein are those of the authors and do not necessarily represent the position of the U.S. Department of Education.