Ellen B. Braaten and Dennis Norman 2006;27;403 DOI: 10.1542/pir.27-11-403

Intelligence (IQ) Testing
Ellen B. Braaten and Dennis Norman
Pediatrics in Review 2006;27;403
DOI: 10.1542/pir.27-11-403
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
Pediatrics in Review is the official journal of the American Academy of Pediatrics. A monthly
publication, it has been published continuously since 1979. Pediatrics in Review is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2006 by the American Academy of
Pediatrics. All rights reserved. Print ISSN: 0191-9601.
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
Intelligence (IQ) Testing
Ellen B. Braaten, PhD,*
Dennis Norman, EdD†
Author Disclosure
Drs Braaten and
Norman did not
disclose any financial
relationships relevant
After completing this article, readers should be able to:
Define intelligence quotient (IQ) and what constitutes the “normal” range of IQ scores.
Describe the predictive validity of intelligence test scores.
Discuss the factors that may influence performance on intelligence tests.
Recognize the relationship between variability observed in factor scores and the
probability of the child having a learning or cognitive disability.
5. Describe how achievement tests are used in conjunction with IQ tests to determine
eligibility for a learning disability.
to this article.
Intelligence tests assess a person’s mental abilities and compare them with the abilities of
other people through the use of numerical scores. Although the term intelligence is used
as if there is agreement on what it means, in reality there is much debate as to how this term
should be and has been defined. For example, debate has surrounded whether intelligence
should be considered an inherent cognitive capacity, an achieved level of performance, or
a qualitative construct that cannot be measured. Psychologists have debated whether
intelligence is learned or inherited, culturally specific or universal, and one ability or several
abilities. While these debates are ongoing, evidence is increasing that traditional intelligence tests measure specific forms of cognitive ability that are predictive of school
functioning, but do not measure the many forms of intelligence that are beyond these
more specific skills, such as music, art, and interpersonal and intrapersonal abilities. (1)
Despite these debates, most experts view intelligence as a person’s problem-solving
abilities, such as adapting to the environment and having vocabulary skills, higher-order
thinking (eg, decision making, reasoning skills, verbal and nonverbal problem-solving
skills), memory, and mental speed. More specifically, for the purpose of this article,
intelligence is discussed as it relates to a child’s score on the intelligence (IQ or “intelligence quotient”) tests that are used most commonly to measure a person’s intelligence for
educational planning or neuropsychological assessment.
Intelligence Tests
Efforts to measure intelligence have long been a part of psychology, and despite controversy over the meaning and scope of intelligence, an IQ score can provide meaningful data
about a child’s cognitive abilities if put within a conceptual framework that does not
overstate its meaning or implications for the child. Intelligence tests are the most studied
and, consequently, the most reliable, valid, and useful tests available for measuring specific
cognitive abilities. Within a particular IQ test, children tend to perform the same on items
designed to assess the same ability, which suggests internal consistency. The tests are
reliable because children generally receive the same score when they retake the same test
years later, although the reliability of the test usually increases with the age of the child.
Test validity is based on numerous studies that have found high correlations between
children’s IQ scores and their performance in school, achievement tests, and tests of
specific intellectual functioning (eg, measures of language, visual motor processing).
An IQ score reflects a child’s performance on an intelligence test relative to that of
children of the same age. In short, a child’s IQ score tells the extent to which his or her
*Psychologist, Massachusetts General Hospital; Assistant Professor, Harvard Medical School, Boston, Mass.
Chief of Psychology, Massachusetts General Hospital; Associate Professor, Harvard Medical School, Boston, Mass.
Pediatrics in Review Vol.27 No.11 November 2006 403
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
IQ testing
test children ages 21⁄2 to 6 years of
age. Each of these tests is composed
of subtests that measure a variety of
domains. The WISC-IV contains
15 subtests that are divided into 10
core subtests and 5 supplemental
(ie, optional) subtests, which form
four composites scales (referred to
as “factor scores”): Verbal Comprehension (verbal knowledge and
the ability to use verbal skills in new
situations), Perceptual Organization (the ability to think about and
Figure. Classification ratings for IQ ranges as they are distributed along the normal curve. organize visual material without the
use of words), Working Memory
performance on the test departs from average. The IQ
(the ability to hold information in memory to manipulate
score represents a construct of “intelligence” that init or perform calculations with it), and Processing Speed
cludes a combination of verbal and nonverbal processing
(the speed at which one can process simple visual inforskills, such as vocabulary, information about the world,
mation without making errors). The Table lists the
reasoning, short-term memory, and speed of information
WISC-IV subtests and factor scores. Although there are
processing; these skills, together, are represented by the
tests of infant “intelligence,” such as the Bayley Scales of
IQ score. Nearly all comprehensive psychological evaluInfant Development, most tests for children younger
ations include some measure of intelligence. For examthan age 3 years measure abilities, such as sensorimotor
ple, for a child who is being tested to confirm a diagnosis
development and early language skills, which are not
of attention-deficit/hyperactivity disorder (ADHD), an
highly correlated with later IQ.
intelligence test can confirm that the child’s academic
difficulties do not indicate a specific cognitive weakness
Predictive Validity of IQ
or mild mental retardation.
Intelligence tests are reasonably accurate at predicting
Most intelligence tests assess a range of verbal, visualwhich children will be successful in school and which will
spatial, and problem-solving skills. Because they target
multiple cognitive skills, IQ tests are composed of
subtests that measure specific areas of functioning. Scores
on these subtests are combined to yield measures of
verbal and nonverbal problem-solving abilities, as well as
Verbal Comprehension Factor
a full-scale IQ score. IQ scores are assumed to be normally distributed in the population, with most scores
● Three Core Subtests: Similarities—Vocabulary—
falling in the middle of the distribution and fewer scores
● Two Supplemental Subtests: Information—Word
falling at the upper and lower extremes (Figure). The
average IQ score on most IQ tests is 100, with a standard
Perceptual Reasoning Factor
deviation of 15. Most IQ scores (about 68%) fall within 1
standard deviation on either side of the mean (eg, be● Three Core Subtests: Block Design—Picture
Concepts—Matrix Reasoning
tween 85 and 115), and almost all scores (99% of popu● One Supplemental Subtest: Picture Completion
lation) fall within 3 standard deviations above or below
the mean.
Working Memory Factor
School-age children most frequently are tested with
● Two Core Subtests: Digit Span—Letter-Number
the Wechsler Intelligence Scale for Children–Fourth
● One Supplemental Subtest: Arithmetic
Edition (WISC-IV). The Wechsler Adult Intelligence
Scale, Third Edition (WAIS-III) is the test used most
Processing Speed Factor
frequently for adolescents ages 16 and older. The
● Two Core Subtests: Coding—Symbol Search
Wechsler Preschool and Primary Scale of Intelligence –
● One Optional Subtest: Cancellation
Third Edition (WPPSI-III) is used most frequently to
WISC-IV Factors and
404 Pediatrics in Review Vol.27 No.11 November 2006
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
IQ testing
have difficulty, with correlations between intelligence
tests and measures of educational achievement averaging
about 0.50. Thus, IQ tests are one of the best single
indices of how well a child will do in school. However,
IQ test scores are not the sole predictive factor of how a
person will perform in school and are not the definitive
indication of how a person eventually will function in
society because other variables, such as intellectual domains not measured by a specific test, parenting, quality
of schooling, motivation, and exposure to culture and
books, also are important determinants of success in life.
Research has shown that IQ constancy increases with
age, although correlations tend to be slightly higher for
elementary students than for high school or college
students. Generally, the correlations with educational
achievement and IQ are highest for verbal subjects, such
as reading. In contrast, the predictive power of IQ test
scores before the first birthday are not very strong for
children who fall in the average to superior range, (2) but
the tests are fairly predictive (ranging in studies from
0.50 to 0.97) for children assessed
at lower IQ levels (ie, below IQs
of 50). (3)
Overall, the general rule of
thumb is that the older the child,
the more stable the IQ. By age 4
years, the correlation with IQ 12
years later is relatively high
(r!0.77). (2) Although many
older children show little fluctuation in their IQ scores, research
has indicated that a subset of younger children show wide
fluctuation in IQ scores. Finally, even older children may
show some fluctuations in scores in response to major
stressors such as a loss of a parent, divorce, or change in
schools. With these possible exceptions, by around age
10 years, IQ scores generally are relatively stable.
education, an enriched language environment, good
school attendance, good schools, and stable neighborhoods. (4)
Cultural and ethnic differences in performance on
intelligence tests also have been documented. For example, studies have indicated that the average scores on
standardized intelligence tests of children from AfricanAmerican and Latino families often are below those of
children from Caucasian families. However, the available
data do not support a genetic interpretation; (5) rather,
the differences likely reflect a cultural or language bias.
In addition to innate and background factors, an
almost limitless list of intervening variables can affect a
child’s performance on an IQ test. A qualified test administrator attempts to minimize such variables as much
as possible, but influencing factors can include the location of the evaluation (eg, noisy office), previous testing
experiences that may result in practice effects, the
examiner-examinee interaction, a negative stance on the
part of the child, peer-group pressure to fail, or poor
Factors That Influence Performance on IQ
In general, children’s factor scores on the WISC-IV
should be fairly similar; the more variability observed in
factor scores, the higher the probability that the child has
a learning or cognitive disability. Previous versions of the
WISC provided verbal and performance IQ scores in
addition to a full-scale IQ. Differences greater than 15
points between a child’s verbal comprehension and perceptual reasoning scores are worthy of an explanation
and may be cause for concern because many learning
disabilities result in large verbal-performance splits on IQ
tests. For example, many children who have dyslexia have
lower verbal abilities compared with nonverbal abilities
because dyslexia is a verbally based learning disability.
.result. .in.largemany
learning disabilities
verbal-performance splits on
IQ tests.
IQ is influenced by genetic factors (eg, the child’s genetic
makeup), familial factors (eg, parents’ IQs and education
and quality of the home environment), educational factors (eg, quality of educational opportunities and teaching), and other factors, such as the community in which
the child lives. Environmental influences on the development of intelligence include access to stimulating or
enriching experiences, caregivers who help the child
learn problem-solving skills, access to books and sources
of knowledge, good nutrition, a high level of social
support, parental involvement in the child’s learning and
motivation. Other causes of poor performance can include limited hearing or visual acuity, a lack of proficiency with the English language, situational stressors,
poor attentional skills, or acute emotional difficulties
such as depression or anxiety.
Discrepancies in IQ Test Score Patterns
Pediatrics in Review Vol.27 No.11 November 2006 405
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
IQ testing
Children who have nonverbal learning disabilities, by
definition, have lower perceptual reasoning scores compared with verbal comprehension abilities and frequently
have significantly weak processing speed scores, as well.
However, even a 15-point difference does not necessarily
indicate the presence of a learning disability. This is
because differences in styles of thinking and learning are
common and often are reflected in a child’s pattern of IQ
scores, such as in the case of a child who has superior
intellect and has a verbal comprehension index of
140 and a perceptual reasoning index of 120. That said,
if an extremely large ("25-point) verbal comprehensionperceptual reasoning split is present, and if one of these
scores is below the average range, psychologists frequently refer the child to a neurologist or to a developmental pediatrician to rule out the possibility of neurologic impairment. Even if the difference between a child’s
factor scores on an IQ test is large, the discrepancy
should not be used alone to make a diagnosis of a
learning disability or to predict brain functioning with-
objective standard. When used to diagnose a specific
learning disability, a child’s academic achievement in one
or more areas is compared with his or her intellectual
abilities. If a child’s ability in one or more areas of
achievement, as measured on standardized tests, is significantly lower than expectations based on age, education, and intelligence, the probability is high that a
learning disability exists. However, these difficulties also
must impede the child’s ability in academic achievement
or activities of daily living. Also, if the child has a sensory
deficit, such as in visual perception, memory, or attention, the difficulties in math or writing need to be worse
than what would be expected with the sensory deficit
alone. For example, if a child who has ADHD has problems with math, the math difficulties must be worse than
what would be expected from a child who has attentional
difficulties. Thus, although IQ and achievement tests are
used frequently to diagnose a learning disability, a simple
discrepancy is not sufficient to make the diagnosis because other issues need to be eliminated.
In addition, the lack of a discrepancy is not necessarily an indication that a learning disability
does not exist. This is particularly
true for the young child who may
exhibit early signs of a learning
disability, but who does not yet
lag behind to the extent that a
discrepancy exists. In these cases,
the pattern of scores on relevant
tests (eg, reading fluency, phonics
skills, reading comprehension, prereading skills) becomes primary in the diagnosis of a learning disability.
Because current federal law recognizes the shortcomings
of a discrepancy approach in determining a learning
disability, school districts are not bound by the discrepancy criteria before children are found to be eligible for
special education services. However, current law also
states that the lack of achievement must not be due to
mental retardation; a visual, hearing, or motor impairment; emotional disturbance; or environmental disadvantage. Thus, intelligence tests typically are given to
rule out the possibility that a cognitive deficit underlies
the child’s difficulties with academic skills.
of IQs with tests
of academic achievement frequently are used
in diagnosing specific learning disabilities.
out substantial support from other test data and observations. Finally, when the differences between a child’s
WISC-IV factor scores are significant, the full-scale IQ
may not be a valid measure of the child’s level of overall
intellectual functioning because the IQ may represent a
forced “average” of very disparate skills.
Using Intelligence Tests to Evaluate Learning
To evaluate specific learning disabilities, such as a reading
disorder, disorder of written expression, or math disability, IQ tests typically are used in conjunction with
achievement tests. Achievement tests are designed to
measure what a child has actually learned, including
mathematical problem-solving, reading, spelling, writing, or an understanding of science concepts. Most
achievement tests focus on a particular subject and measure a child’s learning with questions of varying difficulty.
The child’s score then either is compared with that of a
child of the same age or grade or measured against an
Intelligence is a multifaceted construct that, for the
purposes of this review, is operationalized as the standard
IQ tests (eg, Wechsler Scales) used by schools and psychologists to measure cognitive functioning in a formal
environment. Intelligence scores predict the ease with
406 Pediatrics in Review Vol.27 No.11 November 2006
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
which people learn in formal situations, but do not
necessarily predict success in life or occupations. Measures of intelligence can be affected by fluency of language, access to educational stimulation, educational
resources, motivation, and emotional functioning. Variability in terms of subtest or factor score performance
may be a sign of a learning disability, but a simple
discrepancy between verbal and nonverbal abilities is not
sufficient to diagnose a learning disability. Comparisons
of IQs with tests of academic achievement frequently are
used in diagnosing specific learning disabilities, but
schools and diagnosticians are not bound by the discrepancy criteria. However, to diagnosis a specific learning
disability, the possibility of a cognitive deficit needs to be
ruled out, which typically is done through the administration of an IQ test. Overall, IQ tests are the most
reliable and valid instruments used to measure a person’s
cognitive abilities, but they always should be interpreted
within a conceptual framework that does not overstate its
implications for the child.
1. Gardner H. Frames of Mind: The Theory of Multiple Intelligences.
New York, NY: Basic Books; 1983
2. Neisser U, Boodoo G, Bouchard TJ Jr, Boykin AW, Brody N, Ceci
SJ. Intelligence: knowns and unknowns. Am Psychol. 1996;51:77–101
3. Satler JM. Assessment of Children: Behavioral and Clinical Applications. 4th ed. La Mesa, Calif: Jerome M. Sattler, Publisher, Inc; 2002
4. Spitz HH. Commentary on the contributions to this volume. In:
IQ testing
Detterman D, ed. Current Topics in Human Intelligence: Vol. 5. The
Environment. Norwood, NJ: Ablex; 1996:173–177
5. Brooks-Gunn J, Klebanov PK, Duncan GJ. Ethnic differences in
children’s intelligence test scores: role of economic deprivation,
home environment, and maternal characteristics. Child Devel.
1996;67:396 – 408
Suggested Reading
Braaten E, Felopulos G. Straight Talk About Psychological Testing
for Kids. New York, NY: The Guilford Press; 2004
Flanagan DP, Kaufman AS. Essentials of WISC-IV Assessment. New
York, NY: John Wiley & Sons; 2004
Hebben N, Milberg W. Essentials of Neuropsychological Assessment.
New York, NY: John Wiley & Sons; 2002
Lezak MD, Howieson DB, Loring DW, Hannay HJ, Fischer JS.
Neuropsychological Assessment. 4th ed. Oxford, England: Oxford
University Press; 2004
Obrzut JE, Hynd GW. Neuropsychological Foundations of Learning
Disabilities: A Handbook of Issues, Methods and Practice. San
Diego, Calif: Academic Press; 1996
Snyder PJ, Nussbaum PD. Clinical Neuropsychology: A Pocket
Handbook for Assessment. Washington, DC: American Psychological Association; 1998
Spreen O, Strauss E. A Compendium of Neuropsychological Tests:
Administration, Norms and Commentary. Oxford, England:
Oxford University Press; 1998
Wechsler D. Manual for the Wechsler Intelligence Scale for
Children–Revised. New York, NY: The Psychological Corporation; 1974
Wechsler D. WAIS-R Manual: Wechsler Adult Intelligence
Scale–Revised. New York, NY: The Psychological Corporation;
Pediatrics in Review Vol.27 No.11 November 2006 407
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
cognition, language, learning
IQ testing
PIR Quiz
Quiz also available online at www.pedsinreview.org.
1. Intelligence, as measured by IQ testing, is:
Independent of cultural background.
Invariant over time.
Non-normally distributed.
Not assessable in children younger than age 6 years.
Predictive of school performance.
2. An 8-year-old boy is being evaluated for his poor academic performance in third grade. Results from his
WISC-IV reveal:
Full Scale IQ: 100
Verbal Comprehension Factor Score: 85
Perceptual Reasoning Factor Score: 115
Working Memory Factor Score: 95
Processing Speed Factor Score: 105
These test findings are most consistent with a diagnosis of:
Expected variation.
Isolated attention-deficit disorder.
Math disability.
Mental retardation.
3. A 6-year-old girl whose teachers are concerned about her ability to keep up with her classmates is found
to have a WISC-IV IQ of 70 with a verbal comprehension factor score of 65 and a perceptual reasoning
factor score of 75. These results are most supportive of a diagnosis of:
Average intelligence.
General cognitive deficit.
Isolated attention-deficit disorder.
Isolated nonverbal learning disability.
4. An 8-year-old boy without evidence of sensory deficit is having difficulty reading at grade level. In
addition to a significant verbal-performance split on IQ testing, formal diagnosis of a reading disorder
A full-scale IQ of at least 110.
A home visit.
Grade retention.
Neurologic referral.
Standardized achievement testing.
408 Pediatrics in Review Vol.27 No.11 November 2006
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012
Intelligence (IQ) Testing
Ellen B. Braaten and Dennis Norman
Pediatrics in Review 2006;27;403
DOI: 10.1542/pir.27-11-403
Updated Information &
including high resolution figures, can be found at:
This article cites 2 articles, 0 of which you can access for free at:
Permissions & Licensing
Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
Information about ordering reprints can be found online:
Downloaded from http://pedsinreview.aappublications.org/ at UNIV OF CHICAGO on July 16, 2012