Tuesday, March 15, 2011

The testing dance

In two weeks, our IEP team will meet to discuss a variety of testing the team has been doing to evaluate Joey. When we last met, it became glaringly obvious that completing such evaluations had been overlooked by the school. He hadn't been properly evaluated since he entered the system... six years ago. There is a big difference between a 2-year-old and an 8-year old.

One of the section that had been done was some intelligence and talent testing for the gifted program, because I insisted it be done. I was told he didn't qualify. Yes, he was excellent in math and above-average in reading, but his intelligence testing was only 58%.

Right.

How do you test a child with a communication disorder?

I know there are many "non-verbal" tests out there, but we need to consider how those tests are administered and what they are testing, especially in a child with communication disorders. Showing Joey a picture instead of words and then asking him questions is still testing his disability rather than his ability- because for Joey, it is expressive language and processing questions that causes a problem.

How do you test someone who has a problem answering questions?

For many of us, solutions may seem obvious, as many of my readers deal with their child's communication disorders every day, and we fight to get needed information from them. We know how to win those battles, at least some of the time. And many of us know the tricks: rephrase the questions, make the question into an instruction, try a different activity where the skill must be displayed, use visual cues or language to elicit response, give the child plenty of time to process and respond, pay attention to the responses as a whole. I could go on.

Did Joey's test administrator know these tricks? Did s/he use them? To what extent? How much processing time was permitted? At what disadvantage was he put, and still come out at 58%?

And what do with a twice-special child who has trouble communicating?

10 comments:

farmwifetwo said...

Certain tests must be completed using prescribed methods. IQ testing being one of them. My youngest's IQ score is in the 60's (one year ago), yet his reading/comprehension test they do with all of them came up Gr 3. Why b/c he reads over grade level but answering questions is Gr 1.

There are many tests out there and it is best to ask your psychometrist which ones and why they are going to use.

My eldest has had a couple of full speech and language tests administered over a series of days. The last of these caught the poor short term auditory and visual recall. The psych testing caught the excellent long term memory.

Good testers well note and recognize the child's limitations. Good team members want to know what a child can do, not what the numbers are. Just b/c my youngest cannot ace the IQ exam due to his communication issues, doesn't mean he cannot learn and isn't learning. Sit down and read a book with him. Listen to him talk to the owner of the village grocery store yesterday - yes, he answered her questions; no, he did not make eye contact... BUT he did it, without prompting.

The ACS SLP says he knows the words and has difficulty forming sentences and we need to teach him to build those sentences in response to WH questions. He has the skill since he can manipulate things he hears (tv, computer) and use them as speech.

These skills tell us he's more than his IQ score, where the deficits lie and what we need to teach.

Those are the kinds of tests you need to look for and answers you need to get from his teachers.

Joeymom said...

Its amazing who cares about the numbers. The gifted program people discounted him because of numbers. SOLs are all about numbers. Those numbers should mean something.

And for some kids, they don't mean what the tests intend them to mean. That's a problem if it isn't recognized and accounted for appropriately.

Stimey said...

It's so interesting that you write about this right now. Sam, as part of his recent diagnosis, was tested for this same thing. Now, I imagine you couldn't find a teacher at his school that doesn't have glowing, wonderful things to say about him and his intelligence, yet his overall score was not as stellar as one might expect.

But the private professional we had hired broke the score down into subtests, some of which Sam had scores at 99.5%, some of which had scores at 37%. And even those lower scores, he said would be more accurately represented by a higher score, with holes in it (if that makes sense to you).

All of this is all so complicated. Taking all of the complexities of a child's mind and averaging it out into a percentile is kind of a sketchy proposition. You are so good at showing the school your whole child. Hang in there.

David N. Andrews M. Ed., C. P. S. E. said...

"Yes, he was excellent in math and above-average in reading, but his intelligence testing was only 58%."

Are you sure that wasn't 58th percentile or IQ of 58? because those are the only scores I know of that come from IQ testing: percentiles and scale scores. The difference is very important.

"How do you test someone who has a problem answering questions?
... Did Joey's test administrator know these tricks? Did s/he use them? To what extent? How much processing time was permitted? At what disadvantage was he put, and still come out at 58%?"

These are very good questions, actually. Because those things will naturally affect the scores.

"Its amazing who cares about the numbers. The gifted program people discounted him because of numbers. SOLs are all about numbers. Those numbers should mean something."

Absolutely... they mean something, and their meaning is deeply connected to how they were got. The scores on their own mean absolutely nothing. The numbers, though, are important ... because those numbers (and the details on how they were got) are essential in determining the nature of a child's educational needs and figuring out what to do to support the child's learning. Full-scale IQ scores mean nothing except how many in the norming sample the child scored better than... a meaningless concept that begs the response "So what?!"

They say that the devil is in the details and, in the case of testing, the details that matter most are the subtest scores and the details of how those scores came about.

"And for some kids, they don't mean what the tests intend them to mean. That's a problem if it isn't recognized and accounted for appropriately."

This is absolutely true.

FW2: "There are many tests out there and it is best to ask your psychometrist which ones and why they are going to use."

And this is always a good thing to do - if they can't justify the use of the test to an examinee or the examinee's parent/guardian, then that is a sign that one should run. Bloody quick!

Joeymom said...

Stimey, the "holes in it" makes perfect sense to me. Joey has tons of "splinter skills." He can do advanced things before simple skills people assume you need to do the more advanced stuff. Our best example is that he could say "scissors" before he could say "cat."

I don't know how bad it is elsewhere, but in VA, test scores rule the school. SOLs are everything- and it is those numbers that matter. Tests that ought to be about evaluating what a child can do and getting those gaps filled in are instead just toted up as numbers. Bleah.

Joeymom said...

David, it was 58 percentile (all the testing I know is either in a score or a percentile; we shorthand with the percent mark as many of our therapists and doctors do in notes). This is considered too low to qualify for the gifted program (you have to be at least 80%).

I find that without appropriate accommodation, test scores at school are meaningless. I already know Joey is autistic. I already know he has a communication disorder. You don't have to remind us every year that he needs to be in special education. Can you now tell me what he CAN do and what he DOES know?

The explanation I get for many of the tests being used is that they are "non-verbal", because Joey has a communication disorder. However, when I research the test, I find that every one of the tests are indeed "non-verbal", meaning they use pictures instead of words... but that is not the whole problem with Joey. His problem is more expressive language than receptive language, and processing. If you are asking him a question about a picture and expect an immediate response, you are still testing his disability, not his ability.

David N. Andrews M. Ed., C. P. S. E. said...

"I find that without appropriate accommodation, test scores at school are meaningless. I already know Joey is autistic. I already know he has a communication disorder. You don't have to remind us every year that he needs to be in special education. Can you now tell me what he CAN do and what he DOES know?"

I agree entirely with that. We have the same sort of issues with my daughter still, although she's currently at a much better school now than the one she was in before. My point with scores and so on is exactly that - without any information on how they were got, there's practically sod all use to them, since they give no information in and of themselves on what sorts of supports are necessary for the child to become a more effective learner (which is ultimate what any assessment and intervention process should be aiming at).

"The explanation I get for many of the tests being used is that they are 'non-verbal', because Joey has a communication disorder. However, when I research the test, I find that every one of the tests are indeed 'non-verbal', meaning they use pictures instead of words... "

Indeed. non-verbal tests are not really non-verbal at all: that is pretty much a misnomer. They are visual tests requiring a behavioural output that need not necessarily be verbal.

"His problem is more expressive language than receptive language, and processing. If you are asking him a question about a picture and expect an immediate response, you are still testing his disability, not his ability."

Again, I agree entirely. This is one of the problems for testing when someone has a significant problem with expressive vocabulary (for whatever reason), and it is definitely something that anyone doing a psychometric examination must bear in mind.

In order to test Joey's ability, it would be so much better for the psychologist/psychometrician to follow the test rubric (because those give the results that the tests are normed on) and then (after each item) go beyond the time limits sated by the rubric and see how long the it takes for Joey to actually come up with a response. That would test both his level of disability as well as his ability since he would still be giving responses to the items presented only once (but allowing him time to construct a response in his mind and then vocalise it, or express it otherwise).

David N. Andrews M. Ed., C. P. S. E. said...

(continued)

Sometimes the test rubrics are the problem, and one has to go beyond them to see what really is going on. If that makes any sense. When I have occasion to use psychometric tests, that's what I do when the situation seems to dictate it. Hopefully the psychologist/psychometricians working in Joey's case are aware of this idea too. If not, it might be worth suggesting it. It seems to me somewhat unjust that any child should get low scores on intellectual ability tests purely because s/he takes longer to give a response than the test rubric allows: in a sense, it's a form of structural discrimination (since it discriminates against those who need more time to construct and present a response).

I know that I can - to some extent - speak Finnish; I'm reasonably proficient for everyday purposes ... shopping, getting service in restaurants and the like. But, if some psychologist from Finland were to sue a WAIS-III on me that was normed on the Finnish population, it would form a discriminatory assessment, since I have to take a lot longer to formulate a response in Finnish than I would in English. If you don't get a good response from the psych lot in Joey's team, you can by all means use this example with them just to show how much of a discrimination it is of the test rubric itself prevents Joey from achieving a score he could otherwise get if, say, a time limit were imposed.

It's clear that they are not actually thinking of this, and I'm sorry that this is happening to him. It shouldn't be.

And, yes, the assessment should cover all bases: what Joey does now and can do as well as those things that he cannot do without difficulty.

I'm not sure that they're taking much into account when they test Joey.

Mom said...

Testing for us was a nightmare! Our little guy was 5 when they tried to pinpoint his disorder, but he couldn't even do the tests. At 8 they finally labeled him autistic. He is 17 now and an amazing person. Hang in there! Just don't give up on him. They have their own special gifts.
Hugs,
Sandy
www.twelvemakesadozen.blogspot.com

David N. Andrews M. Ed., C. P. S. E. said...

"But, if some psychologist from Finland were to sue a WAIS-III on me..."

Should be: "But, if some psychologist from Finland were to use a WAIS-III on me..."

Dyslexic-dyspraxic brainfart occurrence ... *sigh*