Healthy Again - but Not at School

I actually feel pretty good today. Whatever my doctor gave me Monday, my sinuses are mostly cleared up and I can actually go to sleep with my mouth closed (which is nice).

I'm hope today, anyway. So is everyone else. We have about four inches of snow at my house this morning. Tazewell County closed last night; McDowell County (along with the rest of West Virginia) closed this morning. We went to bed knowing we had a two hour delay and then got the call at about 6am this morning saying we were closed.

Friday I went to the Board offices for a training session. We have a new IEP form and, more importantly, new requirements within the document. The meeting took about tow and a half hours. I'll spend part of today looking through the new material and browsing the new form.

Every year in March and April we have annual reviews on each special education student and create their individualized education plan (IEP) for next year. I have nine special education students at the moment to think about. Two are leaving the elementary school for the middle school this year, so I don't have to actually write their whole IEP; but I do still have to write their PLEPs (present level or education performance) and send those on to the middle school.

We're also implementing a new program to help struggling readers as part of the response to intervention framework at our school. It's called Recipe for Reading. It falls within the pale of the Orton-Gillingham approached and our county got the institute for Multi-Sensory Education to come in at the end of last year and train us in how to use it.

I'll probably spend part of today looking back over those materials. Our two reading specialists started using it with a few students each on Monday. And I've got a couple of students to start it with, too; but I was out sick...

We'll see whether I get to teach tomorrow. At the moment the weather forecast makes it look quite feasible that we could all be home again for another day.

The Future of IQ Testing in Identifying Learning Disabilities

Note: Visit my education blog, The Green Cup.

The usefulness of IQ testing for identifying learning disabilities is hotly debated. And with changes in the law, issues like cost and ethics are also being discussed.

When reading specialists meet with small groups of students under the new Response to Intervention model, they collect information about how the student responds to instructional strategies and approaches aimed at helping those students "get" what they're having problems with in their regular class. Later, if and when a child's academic problems become serious enough that the child is suspected of having a learning disability, a team of professionals looks at the notes and reports the reading specialist made during those small group times.

Is that it? Is that all there is now to deciding whether a child has a learning disability?

James B Hale, Ph.D. thinks that it is pretty obvious from the new regulations on IDEA 2004 that something beyond just data from the different interventions that get tried with a child will be required.

"So, if you adopt an RTI-only perspective, what multiple measures will be used? How will practitioners assess all areas of suspected disability, including language (i.e., "communicative), motor, cognitive, and intellectual function? ...Maybe I'm wrong, but I think if people are making classification decisions solely on the basis of RTI data, they are now OUT OF COMPLIANCE with the law. RTI data can be *part* of a comprehensive evaluation (as it should be), but it is NOT a SUBSTITUTE for a comprehensive evaluation," Hale said as part of a discussion of the issue on a National Association of School Psychologists listserv.

Dr. Hale is a certified school psychologist and special education teacher who now teaches as part of the graduate program in school psychology at Philadelphia College of Osteopathic Medicine.

The preamble to the new regulations clears up some of the discussion: "The Department (of Education) does not believe that an assessment of psychological or cognitive processing should be required in determining whether a child has an SLD. There is no current evidence that such assessments are necessary or sufficient for identifying SLD...In many cases, though, assessments of cognitive processes simply add to the testing burden and do not contribute to interventions." (p 649). With some 1244 pages of preamble to be waded through, there's probably a lot left to be discovered in the new regs.

In email communication, Guy M. McBride, Ph.D., pointed out that there is a major difference between saying, on the one hand, that we must still perform IQ tests on kids suspected of having a learning disability because the law still (the argument goes) requires it and, on the other hand, saying that we should be performing those tests because they provide us with important information. If the IQ test is no longer a hard and fast requirement, school systems could save money by not testing for IQ - maybe a lot of money when you consider that tests for one child can cost as much as $8,000.

"Depending on the criteria adopted by their states ...public agencies could realize savings under the final regulations by reducing the amount of a school psychologist's time involved in conducting cognitive assessments that would have been needed to document an IQ discrepancy," said McBride. McBride is a school psychologist in North Carolina with over three decades of experience.

Can we use IQ tests to help decide whether a child has a learning disability? If the team evaluating the child thinks that would be useful, certainly. Do we have to use IQ tests anymore? It seems pretty clear from the preamble of the regs that IQ testing is not a legal requirement under IDEA 2004. But §300.304 (Evaluation procedures) of the regs says that "(c)Each public agency must ensure that-- (6) In evaluating each child with a disability under §§300.304 through 300.306, the evaluation is sufficiently comprehensive to identify all of the child's special education and related services needs..."

For school administrators the question becomes one of what it means to be "sufficiently comprehensive" without actually administering the WISC-III (Wechsler Intelligence Scale for Children) or some similar test. And some on the National Association of School Psychologists (NASP) listserv are arguing that the issue is not just legal, but ethical - that it is perhaps impossible to be "sufficiently comprehensive" without a test like the WISC-III, or that it is a disservice to the child being evaluated to not perform such a test.

It will probably be at least a few months before most states announce how they will define "specific learning disability" in light of the new regs - an what role IQ testing will play in the definition...

Response to Intervention, Learning Disabilities, and IQ

Note: Visit my education blog, The Green Cup.

Will we continue testing intelligence? The new Response to Intervention model raises questions about the role of intelligence testing in the future. Do we need it anymore?

School psychologists around the country are among those watching closely as the Response to Intervention model gets implemented. The reason? Intelligence testing has long been controversial and is an expensive component of identifying children for special education services.

"Getting paid for giving IQ tests over the past 35 years has been very, very good to me," said one school psychologist recently in a forum I'll leave nameless. In the context of the larger discussion, to be fair, the remark wasn't as mercenary as it might sound in isolation.

Will we keep giving IQ tests? That question seems to be on the lips of every certified school psychologist, as well as the superintendents who have to pay them and the school principals who have to wait for testing to get done.

Obviously we will continue to give IQ tests when we suspect that a child is mentally impaired. And ruling out mental impairment has been a part of identifying a child as being emotionally disturbed for the last few years; that doesn't look set to change.

IQ tests for learning disabilities? James B. Hale, Ph.D., had this to say after the new regulations were made public: "A quick read over of the regs suggests that after a child does not RTI (respond to intervention), the regs support our position that we should be doing a comprehensive evaluation, including an evaluation of cognitive processes..." Hale is a certified school psychologist and special education teacher who now teaches as part of the graduate program in school psychology at Philadelphia College of Osteopathic Medicine. Hale's remarks were part of a discussion of the issue on a National Association of School Psychologists (NASP) listserv.

Guy M. McBride, Ph.D., is a school psychologist in North Carolina who disagrees with Hale and said so as part of the NASP discussion. "A school may (not must) include cognitive testing but it is not a requirement...," said McBride. "Those who argue that a comprehensive evaluation should always include cognitive testing... are welcome to make those arguments." But McBride believes that under the new regulations individual school districts will make that decision. He says those districts, "would only be obliged to always administer cognitive tests under the LD category if the agency believed it was always appropriate to do so."

The disagreement is easy to explain: the new regulations are not exactly crystal clear in their meaning. The section entitled Evaluations, Eligibility Determinations, Individualized Education Programs, and Educational Placements begins with §300.300 on page 1429 of the new regulations. In the 24 pages that follow we learn that:

• Parental consent is still required in order to evaluate a child to determine if they are, in the words of the policy, "A child with a disability." That means that normal classroom interventions, like having a child spend half an hour a day for a week in a Tier II session on conjugating irregular verbs might get looked at in the evaluation process, but those sessions don't constitute an evaluation process in and of themselves. It's worth noting that once the school obtains permission to evaluate a child to decide if the child has a disability, there is still a 60-day timeline in place for making the decision.
  
  
• The evaluation process must "Use a variety of assessment tools and strategies to gather relevant functional, developmental, and academic information about the child" and must "Not use any single measure or assessment as the sole criterion for determining whether a child is a child with a disability."
  
  
• Whatever might be involved in deciding that a child is mentally impaired or autistic, there are additional procedures for identifying children with specific learning disabilities. And while each individual state will have to come up with guidelines for determining what constitutes a learning disability in that state, the new regs make it plain that states "Must not require the use of a severe discrepancy between intellectual ability and achievement for determining whether a child has a specific learning disability" (emphasis added) and "Must permit the use of a process based on the child's response to scientific, research-based intervention."

Many educators and observers seem to be operating under the misunderstanding that classroom interventions will be the process for identifying children with learning disabilities - and that we'll never have to give another IQ test when we suspect learning disabilities. Judging from the new regs, they're wrong. But the idea that a formal and complete process of IQ testing will still be required appears dubious, and appears to be left up to individual states.

There's much more to say on the issue. And we'll look more closely at the question soon...

Response to Intervention: the Pros

If your child's IQ is between 76 and 90 (that's between 15% and 20% of all kids) the Response to Intervention model probably brings them help they weren't getting before...

It's called the "dull-average" range - not a very colorful or flattering description. One author calls it the "special education no man's land." If a child's IQ is about 70 or below, they can get help from the special education people at a school by being identified as mentally impaired (though there are a few factors other than IQ that enter in to the decision). If their IQ is about 85 or above and they're not doing well in class, they can possibly get help from the special education people at a school by being identified as having a learning disability.

"About" gets defined a little differently from state to state, but basically it means that educators acknowledge that the tests we use to measure IQ aren't perfect. So "about" equals a five point margin for error.

Strangely, when evaluating a child we suspect of being mentally impaired, "about" means a five point margin of error in favor of identifying the child; we'll agree that a child with an IQ as high as 75 can be considered mentally impaired (the cut off for the mentally impaired placement category was 70) if the other factors are in place. But when we suspect a child of having a learning disability the knife cuts the other way. Evaluators often won't agree than a child with an IQ score of 85 has an IQ of at least 85; the child has to score a 90 on the test before the psychologist will nod their head and agree that the child has an IQ of at least 85 (the cut off for the learning disability placement category). When evaluating a child we suspect of having a learning disability, "about" means a five point margin of error against identifying the child.

No one would ever put it this way to a parent (I hope), but under the discrepancy model that's been in use for the last 25 years or so, if Johnny's IQ is between 76 and 90 he's too bright to be mentally impaired and too "dull" to have a learning disability. Johnny's problem in class (the policy's idea, not mine), is that he's just living up to his potential...

Response to Intervention (RtI) takes a different approach. IQ ceases to be the issue in identifying learning disabilities. And the emphasis on identifying learning disabilities is replaced with a focus on preventing them. RtI is partly a reflection of a greater commitment to the philosophical ideal that all children can learn. When children aren't learning, we intervene to help them. And we assume that the problem is the teaching, not the child, until we can prove otherwise.

"Dull average" Johnny gets help that wasn't available to him before; he gets it in the Tier II level of his school's intervention model. Maybe he gets a lot of help - in short spurts throughout the year. And maybe he even gets identified as having a learning disability, though he wouldn't have before under the discrepancy model (and there is still some confusion over that possibility).

The focus on prevention and intervention instead of on identifying disabilities is a positive change (provided identification isn't ignored or avoided as a possibility). And the fact that children will get help based on educational need instead of disability category or socio-economic status alone is an improvement.

There is another potential benefit to the RtI model: cooperation. There has been a divide, a chasm at times, between reading specialists and special education personnel in a school. Each have different mandates and are funded (at least partly) by separate federal programs. Often the reading specialists simply doesn't work with children that have been placed in special education and the special education personnel at a school only take responsibility for a child's education after the child has been determined to be eligible for IDEA services.

RtI creates a school environment where a special education teacher can sit down with a child that hasn't been "placed" yet and provide instruction as part of a Tier II (or Tier III) intervention; as much as 15% of a school's special education funding can be spent in this intervention setting, working with kids that haven't yet been identified for special education services. And the Title I reading specialist can sit right there with the special education teacher and work with the same kid (regardless of socio-economic status) in the intervention framework.

The synergy that could result from cooperation between those two programs could be a strong positive force in a school. And the possibility exists that learning disabilities people and reading people will work together to successfully address the single most common specific learning disability in the schools today: dyslexia.

Response to Intervention: the Cons

Note: Visit my education blog, The Green Cup.

Can we still measure whether a child has a disability?

We talked last week about the new model for identifying learning disabilities: Response to Intervention.

Who decides if Johnny has a disability? In the past it was mostly math. Parents and educators could point to test results and say a child DID or DIDN'T have a disability.

One of the biggest questions the Response to Intervention (RtI) model will face is a pretty simple one: "What IS response?" The discrepancy model, for all its shortcomings, was quantifiable.

You can hear the psychologist (or whomever) at the eligibility committee meeting: "Susie has this much of a discrepancy between her intelligence and her achievement scores, Ms. Smith, and ...well... that means she not learning disabled..."

But can you hear the school's reading specialist making the explanation instead? "Ms. Smith, we've been giving Susie some extra help during intervention time now for about 15 weeks. I know she doesn't seem to be catching up; but when we move her into a one-to-one setting with a specialist and tailor instruction just for her she responds well enough that, in my professional opinion, she doesn't have a learning disability."

Who does decide if Johnny or Susie has a disability? At the moment it's not completely clear whether interventions (and the response Johnny or Susie has to them) are also evaluations, or part but not all of the evaluation process, or simply part of the pre-referral process. Does the fact that Johnny didn't respond to interventions mean in itself that he has a learning disability, or does it mean he should be evaluated to see if he has a learning disability? And if it means that he should be evaluated, then what constitutes an evaluation? The answers to those questions will probably vary from state to state...

It is clear that RtI is intended to replace the discrepancy model as a way of identifying learning disabilities. And it is clear that states cannot require local school districts to use the discrepancy model. It is not clear when the actual evaluation process starts -- when we stop trying to simply intervene and catch Johnny up in reading, and when we start trying to decide if maybe Johnny has a disability. And the new rules will probably allow more flexibility of differences in practice from state to state and district to district than ever before.

In the past it was mostly mathematical. IDEA (the Individuals with Disabilities Education Act, amended in 2004) is first and foremost a civil rights law, guaranteeing the rights of children with disabilities to an education. It relies on the best features of our adversarial court system to ensure that a child's rights aren't denied. And now the only thing that really seems to determine whether a child has a learning disability is the one thing the courts are reluctant to second guess - the professional judgment of educators.

Will it really be a problem? Consider this...

The estimates vary, but somewhere between 50% and 80% of students identified as having a learning disability are dyslexic. Reading is the process of associating a group of mental units we call phonemes with symbols we call letters. Dyslexia is a condition that alters the way sound is processed so that a student with dyslexia doesn't recognize those phonemes the way other students do. What do students with dyslexia do? They cope. They make an effort to associate whole words with the way a group of letters looks to them. Or they employ some other strategy. And they make progress. Some progress. And the more attention and encouragement the child gets, the more progress they make. The problem is, while it is enough progress to hold out hope for those concerned about the child's performance, it is almost never enough progress to catch up. The result? A dyslexic child can appear to be responding (to some extend) to an intervention when in fact we've only slowed the pace at which they are falling behind. And unless their dyslexia has been identified, the chances are that they will hit the same wall in reading (the expansion of vocabulary) that they would have under the discrepancy model - and hit it at the same time (about fourth grade).

Students with dyslexia DO respond to intervention - just not enough to catch up. And the RtI model will then have to decide whether they have a learning disability based on partial response. It is not just an either/or model - though you could get that impression from the simplistic nature of many discussion of the issue.

We've talked around a bigger question. Is dyslexia a disability? The International Dyslexia Association says "yes." I think the solution often has been to ignore the possibility that a child even has dyslexia - despite the fact that there are specific interventions that could be used to teach reading to learners with that disorder.

The success of RtI will be based in part on whether educator in America come to grips with dyslexia - identify it early and address it specifically with interventions designed for dyslexics. If that doesn't happen, my prediction is that RtI will not be a successful approach to learning disabilities any more than the discrepancy model was.

Response to Intervention - A New Model for Identifying Disabilities

Note: Visit my education blog, The Green Cup.

When the regulations for IDEA 2004 became official not long ago, a new model came to town for identifying students as having a specific learning disability. It's called response to intervention, and is abbreviated as RTI or RtI.

To understand RtI and its implications fully you need to understand the previous model for deciding whether or not a child had a learning disability. That model, called the discrepancy model, compares IQ (on something like the WISC-IV) and achievement (usually measured with something like the WIAT) and examines the difference. The idea until now has been that if a child has the intelligence to learn at a particular level but isn't, that discrepancy between intelligence and achievement is evidence of a disability.

The problem was that it takes time, years in fact, for a discrepancy to become significant enough to be considered as a disability. In first grade there's not much to measure. And in second grade there's still not usually a big enough difference between intelligence and achievement to make the discrepancy model work.

When a teacher first suspects that a child may have a learning disability in the first or second grade, everyone felt trapped under the old discrepancy model. The child may well have had a disability, but we needed evidence of the disability in order to place the child in special education. So instead of getting help then, the child got watched more closely. It was kind of like having a rule that you can't throw a life preserver to a kid in the pool until their head goes under the third time. This year the child is just behind, next year the child will have a disability...

In contrast to the old discrepancy model, the RtI model begins looking for curricular intervention designed to catch the kid up as soon as they begin having problems - back in first or second grade. RtI has the potential then to allow disabilities to be identified and defined based on the response a child has to the interventions that are tried. At the very least, RtI can replace the pre-referal process for special education. And in theory, children with real disabilities could be identified and placed years earlier and children without disabilities won't be allowed to fall further and further behind simply to see if they have a learning disability.

The Reading First people have fleshed out the concept of RtI so that most educators today think of a detailed set of specifications when someone mentions RtI. Dallas Reading First has a website that describes the process of intervention well:

• Students get taught good stuff in a setting known as Tier I. This is the primary (or "core") instructional setting for all students.
  
• Sometimes a particular student needs more. So instruction gets supplemented with an "intervention." This is called Tier II. Students come and go in Tier II instruction for short periods of time to address specific areas where the student is falling behind or failing to "get it." Tier II usually involves 30 minutes of extra instruction a day in the content area of intervention. Usually there are no more than four or five students in a Tier II intervention session.
  
• When a student doesn't seem to benefit from 30 extra minutes in Tier II they get moved on up the intervention chart and start getting an extra hour a day (usually) at Tier III. Tier III interventions tend to be more individualized and involve one-on-one time with a specialist.

The Dallas Reading First site describes an intervention plan specifically for reading. And RtI is almost always thought of as a response to reading problems at the moment; but there is no reason the model couldn't be adapted to math or other content areas.

The Dallas Reading First website illustrates almost exactly what former International Reading Association President Richard L. Allington complained about in his article Research and the Three Tier Model in April of 2006. Students in Tier II in Dallas don't get help with what they did during Tier I instruction; instead they get more and different material on reading: they didn't understand McGraw-Hill so they get time in Voyager. But as Allington points out, that's not really a problem with the three tier model itself as much as with the manner in which the model is used.

Will RtI work? I suppose the more salient question becomes who will the RtI model work for? Will it work for the people who think it takes too long under the discrepancy model to identify a child who has a learning disability? Will it work for county administrators and other groups who sometimes feel that too many children are identified as having learning disabilities? And will it work for kids who don't really have a disability but do need help?

I'll take a look at some of the pros and cons of the RtI model in the near future...