stanford binet or WISC?

Post Reply New Topic

Which is a more valid instrument for AS?

I don't know of any research to validate testing with either method to autism specifically. However, even though it is WAAAAAY more expensive, I would say the Stanford-Binet is the better instrument because it has both a higher ceiling and lower floor, so it's better used for MR, LD's, and giftedness.

It's a lot longer though. And most psychologists dread having to give it. There's an entire BOOK that comes along with it, telling the psychologist how to score the whole thing. Not as easy as the Wechsler.

I vote for neither. There is some evidence that Raven's Progressive Matrices produce more accurate evaluations of those with autism. Wechsler is probably what you will get most places though.

I want to be tested with the Stanford Binet L-M because it is verbally biased (I have NLD so my nonverbal deficits drag down my combined score) and the ceiling is much higher. I hit the ceiling of the WISC-III, so I don't know what my verbal IQ actually is.

When you asked about the Stanford-Binet, did you mean the L-M version, or the SB-5? The L-M is really outdated, but it is the only version currently that really has the high ceiling you need to test gifted kids (without extrapolation). The SB-5 is the latest version, but scores above 160 are extrapolated.

I would be hesitant to use the Raven alone. Despite what the Canadian team think, I really don't think it's as good at testing global intelligence as is proffered.

I really think it has a visuo-spatial bias. So instead of using language as the medium to test general intelligence, it uses visuo-spatial abilities to do so.

In any case, it is a poor idea to use only a single instrument to measure intelligence. In the case of autistics who have poor communication, there's little other choice. But for those who have adequate or above average language, then a variety of tools should be used.

That's Basic Psychology Testing 101.

I'm not as familiar with the different versions of S-B, so I'll assume L-M is probably better (if it's available that is).

Well, I assume that when someone with autism is being evaluated, they'll be given a whole battery of tests. The reason I recommended the Raven's Progressive Matrices over the other two is simply that a lot of kids with autism are falsely branded as mentally ret*d when given other IQ tests because of the response mode required. So, RPM can help avoid underestimating what an autie can do, which can be a real problem, especially for nonverbal auties. Although, for an Aspie with stronger verbal than nonverbal skills, the Stanford-Binet L-M may produce good results because of its verbal bias. It's rarely given though, except by psychologists who specialize in evaluating the gifted, because it is outdated. It's main (sole?) strength really is the wide range of abilities it can evaluate. The Wechsler tests and the SB-5 are more balanced because they separately evaluate both verbal and nonverbal intelligence, so I guess they present a more complete picture. The Wechsler at least though relies heavily on crystallized knowledge in the several of the verbal subsections, so that's a real disadvantage to those with autism, who may have had an unusual educational history. I like RPM because of the focus on fluid intelligence. Bottom line of course is that the best option is to give a full battery, although with more prolonged testing you risk fatigue and boredom. You don't want to put the kid through eight hours of testing, even if it is broken up. Sometimes it can be helpful to give portions of different tests, but that has its own problems. When I was tested as an adult, I was given:

From the WAIS: Information, Similarities, Block Design
Wechsler Test of Adult Reading
California Verbal Learning Test-II
Controlled Oral Word Association Test
Brief Visuospatial Memory Test-R
Rey-Osterrieth Complex Figure Test
Judgment of Line Orientation
Benton Facial Recognition
Vigil Continuous Performance Test
Delis-Kaplan Executive Function System- Trail Making
Finger Tapping Test
Grooved Pegboard Test
Tactual Performance Task

That took about 3-4 hours. When you're doing private testing you have the luxury to mix and match like that. In schools though, they prefer having full scale scores to look at, which can be summed up easily for parents and other professionals reading the evaluation reports. So, in order to avoid having the school throw your kid away in a class too low for them, I would recommend the RPM, in addition to the more traditional Wechsler tests.

Last edited by LostInSpace on 13 Jan 2008, 4:21 pm, edited 3 times in total.

Is there a test that is similar to the Wechsler ones, that has a performance and a verbal test, but is harder? Or is the only alternative to use various harder tests and combine the results to get an accurate idea of the verbal and the performance IQ?

Well, although using several tests can be helpful to look at different skills, you can't produce an actual IQ score from combining separate tests, so if you want an actual score, you'll have to use the results from single tests.

In answer to your question- not really- or at least, not that I know of. Raven's Progressive Matrices are better for nonverbal intelligence. Stanford-Binet L-M is verbally biased. And the Wechsler tests or the SB-5 have ceilings too low to differentiate between people further than 3 standard deviations from the mean. Their reasoning (I can see their point) is that there is no point in including questions which can only be answered by (at most) less than 0.1% of the population. Plus, who would be qualified to create such a test? How would they get an appropriate sample to norm it on? What would they base it on? Plus, there is evidence from the L-M that at the highest (and I guess, the lowest) levels, the numbers do not conform to the expected probability curve.

Here's a good article on the use of the L-M in gifted testing:

http://www.hoagiesgifted.org/dont_throw.htm

three SD from the norm would be 145--this is genius level--who cares if a child scores higher than this. I suppose if the child topped off at 145, then one could pursue a different test with a higher ceiling. Let's not forget, folk, that 1% score at this number. It is rare despite our convictions that we and our kids are all geniuses.

MY son scored almost 20 pts lower on the SB-5. He originally took the WISC and scored in the gifted range (it was not up in the 140's--believe me).

I think we know if someone is a genius. It is apparent. My uncle was a genius. He was also manic-depressive and a great poet and artist. But, you didn't want to have a conversation with him because he stared at you strangly, said odd things and was intimidating to say the least. Yet, I was always mezmerised by him and he had a following because he was so knowledgeable (he was also extremely religious). I wish I had known him better--he passed on.

My point being, I believe he was a genius because he was intense and different. A genius isn't working 9-5, and reading the Sunday paper in his/her recliner with a beer or glass of wine. They are too complicated, I imagine, to live such an average life. I'm not a genius myself, but I am creative, write fiction, and imagine what one thinks. It's the best I can do. I struggled to get a B in math and all else that wasn't my interest at the moment. I

I think there are quite a few pseudo-geniuses.

equinn

Well, my verbal IQ has been measured at 147- that's the 99.9th percentile (or rather above it- the WISC stops providing normative values at 145). That's using the "gold-standard" of IQ testing, the WISC (not an internet test, in case you were wondering). And it was measured 2 separate times almost ten years apart so I think it's probably accurate. However, I do *not* consider myself a genius. I'm smart, but I spend my time much like anyone else (and I wish, like everyone else, that I had more time to explore what truly interests me). And I've known a number of people I consider much smarter than myself- people who might truly be geniuses. 145 is just not high enough of a ceiling to identify kids who differ from the norm enough to require intervention. Consider this quote:



There are profoundly gifted kids out there who may not be able to thrive within a traditional educational curriculum. We'd be failing those kids, just as much as we'd be failing kids with learning disabilities, if we neglected to accommodate their special needs. Kids with an IQ of 180 can do poorly in classrooms because what they are being taught is so below them that they fail to engage. It is important to identify those children so that they can be given *appropriate* educations. I managed within a regular classroom, although certain subjects were frankly complete wastes of time. Someone much smarter than me (who would not be differentiated from me using the most common IQ tests) would have faced many more obstacles. Here's another quote about profoundly gifted children:



Lewis Terman, who developed the first ever IQ test, said "The child of 180 IQ has one of the most difficult problems of social adjustment that any human being is ever called upon to meet."

Very true, I agree. The only thing though is that I gathered from Michelle Dawson that their team is proposing it as a single method of IQ testing. That was the greatest part I disagreed with her on.

Perhaps she explained it incorrectly or somehow I misread her. But I felt I read her fairly clearly in that it should be THE test to use. And not just for autistics, but for the general population as well.