A- A A+

Putting the Discrepancy Model to Rest

Testing for Learning Disabilities - Children with Learning Disabilities

Intelligent Thinking vs. Intelligence Testing

Much has been written during the past three decades about approaches to evaluation and decision-making regarding who does and does not qualify for the classification of specific learning disabilities (LD). The vast majority of students who have undergone assessments to determine eligibility under the LD category have been subject to a battery of tests, most often administered by individuals who had:

  • no first-hand knowledge of their instructional experiences in school
  • limited access to parents and subject-area teachers
  • some information about scores of different sorts (report cards, standardized assessments)
  • little or no qualitative and quantitative information about performance on subject area tests, projects, homework and participation classroom discussions

I believe that a combination of common sense and the momentum behind the Response-to-Intervention movement is beginning to have a dramatic (and positive) effect on how we think about LD and how students who struggle with learning receive the instructional and behavioral supports and services they need to be successful learners. Of the many still unresolved issues related to LD identification, one seems to loom larger than others. Simply stated, what role, if any, should intelligence testing play in the determination of LD status?

A Problem of Definition

The answer to this question may seem simple, but it’s not. Consider the following:


Consider This

By definition, individuals with LD are of at least average intelligence, and the only way to determine whether a person is of average or above average IQ is to administer a formal test of intelligence. If a fifth grader who reads the newspaper every morning (with good comprehension) and who is an avid chess player struggles to complete homework assignments on time and does poorly on multiple choice tests in Social Studies, is an IQ test needed to rule out low IQ?
IQ tests can be useful tools as part of a comprehensive LD assessment, and data gleaned from IQ tests are useful for planning interventions and making instructional decisions. What about the ninth grader who is succeeding in accelerated Math and Science classes but is ‘underachieving’ in school because of poor reading and writing skills? How will proof of an average or above-average IQ contribute to instruction planning and support services?
A discrepancy between a student’s IQ scores and performance on measures of academic skill (e.g., reading, math) has been the hallmark of LD for the past thirty years and continues to be a valid approach to LD determination. Dozens of models have been used over the years to compare scores on assessment tools and determine eligibility for the LD classification. With states, districts and even individual schools abiding by different cut points, and using different formulae or criteria and allowances for “professional judgment” to tip the scales in the absence of convincing data, it’s hard to think of a discrepancy approach as having any meaningful validity.
Psychological processing deficits are essential elements of the LD determination process. Learning efficiency, attention, memory, meta-cognition and social cognition are all important factors to be considered during the LD determination process. IQ testing per se does not address any of these, except perhaps by inference.

New thinking about these assumptions (and others) together with IDEA-approved opportunities to use alternate approaches to LD determination (long overdue and welcome changes in the status quo for both general and special education practitioners) hold enormous promise to providing more timely and effective help to all students who struggle with learning, including those with LD.

Where does IQ testing fit in? Should uneven performance (after well-targeted instruction) and repeated reports of unexpected underachievement in real learning situations be enough to qualify students as having LD? Can assumptions made about IQ, at least for some students, be sufficient to rule in ‘average intelligence’ for the purposes of LD classification? Read on.

Is IQ Testing Really So Important?I did a quick, informal survey of fifteen teachers and parents in some neighborhood schools. When asked the question “How important is IQ testing for determining LD in school-age students?” adults offered comments that directly reflect the controversy about whether IQ testing is meaningful to the LD determination process:

Adults who admit to having had no experience with or knowledge of IQ testing said...

  • “If IQ tests are going to help teachers pick out the kids who are the best learners and others who are not as smart, that should help them give more attention to those who need extra help.”
  • “Most students are not going to have IQ problems. And I think that schools did away with IQ testing when I was a kid. If a kid needs testing, he should have it, but the school has enough to do with kids who have serious problems. I'm not sure what LD has to do with IQ testing... I thought kids with LD had dyslexia and hyperactivity disorder.”


Adults who admitted to having some knowledge about IQ testing through a friend, neighbor or sibling said...

  • “No one ever talks about the IQ test or gives out scores, so I’m not sure what the test really offers.”
  • “IQ tests are important to prove that kids are smart enough to learn. If they have serious IQ problems, they probably belong in special classes or at least have modified tests and assignments.”


Adults who, as teachers or parents, had children undergo special education assessment, including IQ testing said...

  • “The different subtests were interesting... some were harder for my son to do than others...but I can’t say what the total score or the sub-scores helped my son’s teachers know what to do.”
  • “IQ tests don’t predict whether my child will learn to read better before he gets into high school. That’s really what I care about. I wouldn’t care if they didn’t use the IQ test because everyone knows he’s bright. They make it into a big deal but it’s really not.”


Food for Thought

While this sampling of responses was totally unscientific and intended for descriptive purposes only, it does lead us to ask a few additional questions:

  1. Is it possible for parents and teachers, in some cases, to infer ranges of intellectual or cognitive functioning, without the formal administration of IQ tests? If so, at what risk? If not, are there reasonable alternatives that might enhance decision-making and instructional planning?

  2. If the ultimate purpose of LD assessment is to help determine better strategies and approaches to instruction for students who struggle with learning, what specific information might tests of intelligence and information processing add to help educators decide what instructional practices are likely to result in improved student learning?

  3. Wouldn’t it make more sense to focus LD assessment activities on understanding specific areas of skill deficit by engaging in well-targeted, intentional teaching, monitoring student progress, and sampling information processing behaviors along the way?

  4. Should any testing for LD be initiated before first ensuring that a student has had adequate instruction in areas of frustration and poor achievement? And last but certainly not least,

  5. Is it time to revisit what we know about LD, what we think we know, what we still need to discover, and;
    • Rethink the current definition of LD?
    • Consider the policy implications of any changes to the LD definition and how it might affect the preservation of the LD category under the law?


Additional Resources

  • Fuchs D. & Young, C. (2006). On the Irrelevance of Intelligence in Predicting Responsiveness to Reading Instruction. Exceptional Children, 75,1,8-30. Council for Exceptional Children. Arlington, VA.
  • Kavale, K., Holdnack, J. & Mostert, M. (2005). Responsiveness to Intervention and the Identification of Specific Learning Disabilities: A Critique and Alternative Proposal. Learning Disabilities Quarterly, 28,1. Council for Learning Disabilities. Leesburg, VA.
  • Mellard, D., Deshler, D. & Barth, A. (2004). LD Identification: It’s Not Simply a Mtter of Building a Better Mousetrap. Learning Disabilities Quarterly, 27,4, 229-242. Council for Learning Disabilities. Leesburg, VA.
  • Scruggs, T. & Mastropieri, M. (2002). On Babies and Bathwater: Addressing the Problems of Identification of Learning Disabilities. Learning Disabilities Quarterly, 25,1, 155-168. Council for Learning Disabilities. Leesburg, VA.


Sheldon H. Horowitz, EdD is the Director of LD Resources & Essential Information at the National Center for Learning Disabilities.