I have been an outspoken critic of IQ testing for children with special needs for many years. I believe, based on my review of the literature and my own experience that school districts consciously or subconsciously predetermine a child's outcome based on his or her IQ. If you have a higher IQ, they may deny needed support because, after all, you can get good grades; if you have a lower IQ, expectations are dumbed down, students participate in "community learning" (you know, that is where they learn custodial skills from the school custodian), and administrators insist that your child can't "get" whatever it is that is taught in the "other" classrooms, where the government demands "high standards" and "accountability" for neuro-typical children.
Okay, IQ testing pisses me off. My son's "first" IQ equivalent was taken by Dr. Doom and Gloom when he was two. She told me he had an IQ less than 55. She did not know how much less but he didn't make it to the lowest cut off. I walked out of there in a daze. A few months later, in the offices of the doctor my son's district liked to refer to as the "Beverly Hills Diva", his IQ measured at 137. Yea, it was a much better number but all I could say to Dr. Diva was "why should I believe you and not the doom and gloom number that I was assured was valid, reliable and that I needed to simply accept".
Well, my education was just beginning. Many children with disabilities simply cannot be tested with traditional IQ tests - at least not while they are severely impacted by their disability. Oh, districts deny this and claim that IQ tests are normed in a way that valid information is obtained but I have seen the variability in test numbers depending on what test is given and who is giving the test. Yep, the variability kept manifesting itself with my own son. The idiot psychologist that was hired to do his kindergarten transition assessment so muddled the oral instructions for my child with an auditory processing disability that the score was in the low 70's. I objected to what she did during the test and ultimately withdrew consent so that the scores could not be given to the school district to wrongfully make conclusions about my son's intellectual ability. When the district insisted that it "had to" assess in all areas of disability, I asked them what evidence they could point to that my son had a suspected intellectual disability. That stopped them, albeit only momentarily. I continued to refuse and they didn't have the guts to file a due process to try to obtain an IQ test, in part, I suspect, because they knew he was not cognitively disabled.
Once my son had undergone years and years of remediation, I did privately assess his IQ, knowing that his district would want to do so. My thought was that if it came up high, I would give them the private testing and then refuse their own people from assessing him. I had both verbal and non-verbal IQ tests done and they all came up in the average to high average range. I gave them the tests and that stopped them from asking any more. Of course, by then, he was fully included on a general education campus, taking AP classes so the testing was simply for the sake of getting the real diagnosis of his reading disability (which districts insist can only be measured by looking at the gap between cognitive ability as measured on an IQ test and academic achievement, as measured on various instruments like the Woodcock Johnson or the WIAT).
So, does the research support my own personal experience? Yep. In 2009, researchers reported that people with autism often struggle with the verbal portions of the Weschsler Adult Intelligence Scale (WISC), the most commonly used IQ test. If given a non-verbal IQ test or even a test of abstract reasoning abilities, such as the test often used to evaluate for gifted programs the Ravens Standard Progressive Matrices, autistic people not only had scores equal to those of their non-autistic counterparts, but they answered the questions, on average, as much as 42% more quickly and, on average, autistic participants scored 30 percentage points higher than would have been predicted by their scores on the WISC.
So, with the research showing this problem, why would districts continue to do the WISC? Well, that is a good question. Maybe they only know how to do the WISC. Maybe that is the only test they purchased. Maybe they believe all autistic kids are intellectually disabled anyway so what difference does it make.
Fundamentally, however, what does IQ get you? If services are supposed to be determined based on a student's needs, what relevance does IQ have? I submit, NONE. In fact, that was the line I gave my son's kindergarten transition team and many other teams ever since. When a parent refuses to consent to IQ testing, I suggest that they tell the district what I told my son's district after revoking consent so that they could not get his IQ test scores. I had consented to to a wide variety of other tests (like speech testing, fine and gross motor testing, academic testing, etc). I told them: "You have my consent for some assessments. Once these are completed and reviewed, I will reconsider my refusal to consent to IQ testing". When they reported our the results of other testing, showing my son to be 2 or 3 years above grade level academically, 2 to 3 years below grade level on speech and language skills, etc., we were readily able to write an IEP. The speech scores showed the primary areas of deficit. The IEP was written and then the director once again asked for IQ testing. I turned to the teacher at the table, who was principally responsible for the IEP, and asked this simple question: "Would you do anything different if you knew my son's IQ was 120?" He said, "No". I said, "Okay, would you recommend anything different if you knew my son's IQ was 85?" He said, "No". I said, "The truth is that regardless of whatever number might show up on some test, you believe, don't you, that this IEP appropriately meets his needs". When he said "yes", it was all over.
Yes, the director tried to argue that "we have to" assess. I corrected her and said, "No, you have to "consider" assessments and only NEED to assess in all areas of suspected disability or to determine needs. And, if you think about it, other than for students with a suspected intellectual disability, what "needs" will be determined with an IQ test? Don't we look at 'needs' by looking at things like academic testing, processing tests (auditory, sensory, visual), speech, gross and fine motor, etc? And, even if there is a suspicion of an intellectual disability, what value does the number have for the child? If we know he has an ID, the IEP team can determine eligibility and the number doesn't make one hill of beans difference.
For autism, it is imperative to remember that eligibility is not dependent upon cognitive scores. Unlike eligibility for an intellectual disability, a student can "qualify" as a student with autism with either a low or high IQ. So, IQ is not necessary for autism eligibility. If the sole suspected area of disability is autism, an IQ is not necessary to make that determination and "needs" can be assessed with a host of other data collection tools that ferret out academic, language, and processing needs. With autism, the research makes clear that you can and often do see significant variability in index scores for subtests that are measuring the same construct. On the KABC-II, for example, the Simultaneous Index includes the Rover and Triangles subtest. Dr. Kaufman, in publishing the manual, advises that students with autism often score very high on Triangles and very low on Rover. If the spread is significant, the index is considered "not interpretable" and can not be used to compute an IQ score. Do you think that would stop a school psychologist? Nope. The last three students who had district administered KABC-II assessments had uninterpretable scores yet each was interpreted by the pscyhologist and no cautionary language was written in the report. The IQ was reported as valid, reliable and interpretable even though the manual made clear that they were wrong. Is this a training issue? Or do the psychologists know and report the IQ out anyway?
School districts who want the IQ number for an improper purpose and know that they don't really need it to determine eligibility or needs, may claim that they want to see the "pattern of strengths and weaknesses". But, if you really are interested in the pattern of strengths and weaknesses, there are much better assessments and ones that will look at each processing area distinctly. The WRAML will look at memory, the TAPS can look at auditory processing, and I can go on and on. The insistence for an IQ test, in my opinion, is because the district really wants a number so that they can say forever more "well, Billy is working consistency with his expectancy". What is amazing to me, is that on some of the students with whom I work and for whom I know an IQ number, they are working well above their expectancy? How can that be? Maybe the number is not really reflective of their expectancy which is exactly what the research shows. Wanna bet that if a parent offered to allow an IQ test but NO computation of the composite score and consideration only of the subtests, the district would INSIST it MUST compute the number? The KABC-II manual does not mandate that and, in fact, it makes clear that if there is significant variability among the Scale Indexes, you should not compute the composite.
The reality is that school districts take IQ tests and pre-determine outcome. Maybe they don't do it intentionally but again, research shows that they do. In a Fall 2005 article in Disability Solutions entitled the "Least Dangerous Assumption", author Cheryl Jorgensen, Ph.D. discussed the problem with teachers subconsciously expecting less from students they THOUGHT were mentally retarded (we now use the word "intellectual disability"). If they think that a student is mentally retarded, they may place the child in an MMR class, not try to teach her phonemic reading skills but only functional sight words, she might spend the bulk of her educational career learning functional skills, and be spoken to in short phrases, as if she cannot understand regular conversation. And they might have been wrong. The IQ test that placed her in that program may have been testing her disability and not her ability.
I can't tell you what immense sadness I had when reading a case a while back about a 20 year old young man who was then in 11th grade in Atlanta. This student had struggled in elementary school and, in 1988, when he was in 4th grade, his district finally evaluated him. The school psychologist determined that his IQ was 63 ? and found him eligible as a student with mental retardation. Based upon his low IQ, his school dumped him in a self-contained class for students with intellectual disabilities. Five years later, his high school attempted to evaluate him again and still concluded that he had an IQ of 60, even though there were significant variations reflecting that the IQ score may not be accurate. Finally, a different school psychologist evaluated the student and concluded that he was NOT mentally retarded but instead had a reading disability. This student was stigmatized with the MR label for 5 years, deprived of any opportunity to learn to read, and was still in school at the age of 20.
So, yea, I do not like IQ tests.
Stories like this caused Dr. Jorgensen in her study to conclude that educational decisions should be based on assumptions which, if incorrect, will have the least dangerous effect on the likelihood that students will be able to function independently as adults. Thus, the least dangerous assumption is to conclude that students with significant disabilities are competent and able to learn, because to do otherwise would result in harm such as fewer educational opportunities, inferior literacy instruction, a segregated education and fewer choices. Another reason for assuming competence is that research is now showing that with high expectations, good instruction, and the support of assistive and communication technology, a growing number of people labeled as mentally retarded acquire literacy skills and demonstrate intelligence beyond what would have been predicted by their tests results (Jorgensen).
From my perspective, if the district can't assess the student, it makes it harder for them to "label" them as a student with an intellectual disability and harder for them to make predetermined decisions about their educational outcomes. Shouldn't we decide what a student can do based upon what they can do and not based upon a number? If we do that, what is the harm? If we follow the reverse process and determine outcomes based upon the number, the risk of harm is significant if the number that was obtained was mis-interpreted (intentionally or negligently) because of significant variability or if it simply is not reflective of the student's actual ability. That is what happened in the case in Atlanta where a child without an intellectual disability was placed in an ID class and lost years and years of opportunity. I have students with autism who, if tested, would demonstrate some level of cognitive impairment but are performing grade level work. They are included on general education campuses. They get support, for sure, but they are not warehoused into functional skills programs. My theory is that it is imperative to keep the expectations high, and keep the pressure on; only time will tell whether or not the child will actually be significantly limited in their outcomes.
I am not the only voice out there that is talking about the risk of lowered expectations. Professor Richard Peterson has researched the problem and concluded that the "culture of low expectations has infected national, state, and local educational agencies and has frustrated society from realizing the veision of the IDEA to improve the results of education for [special education] students so that they may be empowered to participate in and contribute to society, and work toward achieving further education, employment and independent living".
So, this is one of those times where my mantra is: "Just say no".
November 1, 2013
IQ Testing: Should I Say No?
January 11, 2012
What's in a Label? When a Rose By Any Other Name May Not Be a Rose.