Panel 3 Speaker: David Tilly, Ph.D. - Heartland Area Education Agency (IA)
RTI Leadership Forum
December 8, 2010
Thanks Steve. I’d like to open by thanking Kathy Whitmire and Jim Wendorf and NCLD for hosting this meeting. I agree with Doug. I think this is an important meeting and activity and we need to do it more often. My comments today are going to be based on my experience working as a school psychologist actually implementing and overseeing and administering programs that not only identify students with disabilities but also help program for them in Iowa for the last 20 years. I’d like to accomplish two things with my comments today. The first is to place this discussion in something of a context, a historical context and talk about what we’re doing and why we’re doing it. And then secondly I’d like to identify 4 important ways that comprehensive assessments actually change on the ground when you start implementing RTI systems.
Prior to the passage of 94-148, there were over 1,000,000 children with disabilities in America who were denied a free and appropriate public education simply because they had disabilities and public schools were not prepared to teach them. At the time finding children with disabilities and ensuring their rights, ensuring that their rights were respected with both a significant political and social policy outcome that we were all after. So we put a lot of energy into it and have had tremendous success finding and identifying students with disabilities. Witness the rapid growth of special education since 1975. Unfortunately the procedures that have been so successful in finding and identifying children have not been equally successful at accelerating their learning. The vision of IDEA was not that students with disabilities would receive an appropriate program. The vision of IDEA was that that program would be effective. This value is reflected clearly in the last purpose of IDEA, which basically states that part of the purpose of IDEA is to assess and ensure the effectiveness of efforts to educate children with disabilities.
RTI and special education identification has emerged primarily to take aim at adding the effectiveness component to the process that so effectively finds students and protects their rights.
RTI as a framework can be used both to assist in the identification of specific learning disability and importantly to assist in programming for students with disabilities. As Jack mentioned in his paper, a majority of the comprehensive evaluation requirements are consistent across the disability areas and do not change as a result of implementing RTI. There are however a series of ways that comprehensive assessments change in practice when implementing within an RTI system, and I’m going to address 4 of those today.
The first way that RTI improves comprehensive assessments is that it helps refocus assessments away from psychometric assessment towards what Carver in 1974 called “edumetric assessment.” RTI improves our assessments by changing targets for assessments so for example instead of traits and within person variables, we increasingly are measuring skills that those traits impact. RTI improves our assessments by increasing the directness of our assessment. So instead of measuring indirect correlates of learning such as memory, processing, style, sub-test scatter, on measures of learning aptitude using items that are somewhat removed from classrooms, we instead measure student’s skills directly on classroom-relevant tasks often using materials selected directly from a student’s curriculum. Because the time we have for assessing students is limited and finite, we have to focus our time on the variables that are most related to improved outcomes. The rule for assessors is clear: don’t measure correlates when you can measure the criterion.
RTI improves our assessments by allowing us to better also tailor our assessments to individual students. I don’t know if many of you are like me, but when I started practice we used to be involved in administering what we called our standard battery and for us it was 9 things and it didn’t matter what kind of problem came across our desk, we did the same 9 things. Didn’t matter what the suspected disability was, we did 9 things. With RTI we’ve changed our approach from a comprehensive battery to be more in line with a full and individual evaluation. In that case it’s possible to tailor the specifics of an assessment to each child’s needs and their possible disability. So instead of coming up with a diagnosis, a general diagnosis that’s general and with little relationship to classroom teaching, we come up with specific educationally relevant diagnoses. So for example instead of in our old system telling parents that their child had a learning disability based on general tests of aptitude and achievement using a regression formula, we might tell the parent based on edumetric assessments that their child has a specific reading disability and here’s what it is. Your child’s phonic skills have some critical deficits. We might give some examples. She does not understand many of the vocabulary words in her grade level reading material. Again, give some examples. She’s not monitoring meaning while she’s reading. And her reading is very labored and slow. All of these things impact her ability to comprehend what she is reading. Each of these things is directly teachable and her instructional program will focus specifically on remediating these skill areas for the purpose of improving her reading comprehension and ultimately her academic success.
A second way that RTI changes comprehensive assessment process has to do with the two-pronged test required to entitle a student with a disability. Since the late 1980s it’s been clear that for a student to be eligible for special education they both had to be a student with a disability and they needed specially designed instruction and related services in order to receive a free appropriate public education. The federal regulations are clear about this and explicitly spell it out. The regulations do not however specify an order in which these tests must be attended to. In the old system, students were referred by their teachers. When they were referred, an evaluation of disability was completed and, I don’t know if your system’s like ours, but oftentimes the need component was more of a cursory afterthought. Obviously the student wouldn’t have been referred if they didn’t need something different.
In our RTI system, the need prong is examined first through the provision of an evidence-based instruction in general education. Progress is monitored and decisions are made based on how children are learning over time. If a student’s learning can be supported successfully in general education and the student does not need specially designed instruction in order to receive FAPE, then the student will not qualify for special education. The student could have an IDEA-eligible disability but they don’t qualify because they don’t need the service.
A third difference has to do with the way we’re looking at problems for students with disabilities when we implement assessment within an RTI model. When examining student learning problems with RTI we can look not only at the intensity of student problems—the summative dimension—but also problem severity which combines both the summative and formative dimension of student problems. Perhaps an analogy will help. If you have a very high temperature, say 103 degrees, and you go to an emergency room you have an intense problem, a difference between what is expected, 98.6, and what’s occurring. However, you do not yet know how severe your problem is. If in the emergency room they treat you, say by giving you Tylenol or some such medication, then they monitor your progress over time to see how you’re responding, those data tell medical professionals important information. If your temperature improves markedly over time, your problem was intense but not severe. On the other hand, if after treatment your problem persists or gets even worse, your situation is more severe which has implications for further treatment. The same type of analysis is relevant for students with developmental learning problems. If we can identify problems early and provide appropriate instruction early, we can distinguish between children whose problem is instructional mis-match and those with disabilities. That’s something our historical system did not do very well. This is an important increment to our assessment system.
Finally, implementation of RTI in schools moves us away from an a priori aptitude by treatment model, aptitude by treatment interaction model, to a post-hoc aptitude by treatment interaction model. What I mean by that as we’re all aware, an ATI model suggests that we can measure characteristics of an individual and match effective treatments to those characteristics. The idea sounds logical. It ought to work, and it works in a lot of human service professions like medicine. However, an education, whether we’re talking about measuring aptitudes, patterns of strengths and weaknesses in cognitive processing or learning styles, we have over 30 years of research that says it just doesn’t work. Indeed as Lee Cronbach reminded us in 1975, once we attend to interactions we enter a hall of mirrors that extends to infinity.
There are very few universals in education, but it appears that at our current state of knowledge this just might be one of them. We can’t predict with certainty whether a particular instructional strategy or package of instruction will work for an individual student prior to trying it. As such we have to move away from an a priori or before-the-fact model to assumptions of a post-hoc or after-the-fact model. We have to give it our best shot. We need to use our best technologies to assess, take our best guess about what might work for a student, implement the program, and then measure, monitor and adjust.
To close my comments, I guess I’d like to say that RTI wasn’t created for the purpose of disability identification. While it wasn’t created for that purpose, the use of RTI models has improved on the ground assessment practices in many schools across America. It doesn’t solve all of the conceptual or measurement problems associated with accurate identification and diagnosis, and it will not guarantee academic and social emotional success for each and every child with a disability. When used appropriately, however, it will provide teachers and parents with high-quality instructionally relevant data that can be used to develop and implement instruction with a reasonable likelihood of success. And I know that every person in this room agrees with me when I say our children deserve no less. Thank you. (applause)
Back To Top