Duke TIP

IQ Tests and Gifted Children

August 28, 2006

Measures of IQ, the intelligence quotient, have historically been the primary means of identifying intellectually gifted children. Although additional identification methods have been adopted in recent decades, IQ tests retain a prominent role in determining appropriate educational settings for highly able students. The following is a brief introduction to IQ tests and IQ testing.

A Bit of History

Modern intelligence testing is almost as old as psychology itself, that is, a little over 100 years old. The origins of the discipline of psychology can be traced to research conducted by physiologists and neurologists in the late 19th century. A wide variety of tasks were investigated as possible measures of IQ, and those that appeared to work well were developed further. By the outbreak of World War I, the idea of IQ as a ratio of mental to chronological age was established. Using this definition, a six-year-old child who performed like the average eight-year-old was assigned an IQ of 8/6 times 100, or 133. Although this definition has intuitive appeal, it is no longer used today, because it cannot be applied to adults.

Many IQ tests have been developed. Commonly used intelligence tests include the Stanford-Binet (S-B), the Wechsler Intelligence Scale for Children (WISC), the Kaufman Assessment Battery for Children (K-ABC), and the Woodcock-Johnson Tests of Cognitive Ability. Tasks are broadly similar across tests, with some variation that depends on the tests’ theoretical conceptualizations of intelligence. For example, most IQ tests contain some tasks that draw on vocabulary and others that require pattern recognition. These tests are usually administered individually, rather than in groups, by a licensed psychologist.

A Bit of Theory

All IQ tests are designed so that an average IQ is represented as a score of 100. The spread of scores is measured by the standard deviation. For historical reasons, the standard deviation on most IQ tests is set at 15 (sometimes 16). Most people score near average (100), and a few have either very high or very low scores. This pattern is called a normal distribution, better known as a bell curve.

Image: IQ Bell Curve

Giftedness is typically defined by scores at or above 130. About 2 percent of individuals score in this range. In some locales, an IQ score of 120 is used for entry into gifted programs. This broader criterion includes about 10 percent of students as gifted. A single school, or even a single district, may have many more or fewer students with IQ scores at a given level, because the normal distribution is based on a very large population and a particular school may not be fully representative of it.

The highest possible score is known as the test’s ceiling, and students whose abilities fall at or above this level receive the same numerical rating even though their abilities are not identical. This “ceiling effect” often poses a problem for highly gifted students, because many standardized tests (including most intelligence tests) are designed to work best within three standard deviations of the average score. The closer a score comes to the test’s ceiling, the less accurate it is as an indicator of the child’s level of ability. Therefore experts recommend that highly gifted children be administered intelligence tests with a high ceiling. One intelligence test that is sometimes recommended for this reason, even though it is outdated, is the Stanford-Binet, Form L-M. An experienced psychologist is aware of the issues that each particular test raises and can explain to parents the rationales, for selecting one test over another for the purpose of assessing their child.

Test Name Version Ages

Stanford-Binet

L-M
IV
V

2 years to adult
2 years to adult
2 years to adult

Wechsler Intelligence
Scale for Children

III

6 years to 16 years, 11 months

Kaufman Assessment
Battery for Children

II

3 to 18 years

Woodcock-Johnson
Tests of Cognitive Ability

III

2 years to adult

Additional Advice

For parents who wonder what should happen when a psychologist tests their child, the American Psychological Association publication Standards for Educational and Psychological Testing (available at www.apa.org/science/standards.html) offers clear guidelines that testing psychologists follow.

Nearly all modern IQ tests have been developed in accordance with strict standards for technical adequacy and are quite suitable for determining whether your child meets local criteria for entering a gifted program. They also provide information that can support your advocacy for appropriate educational services for your child. You can learn more about these and other tests in the Buros Mental Measurements Yearbook. This annual yearbook provides reviews of tests and is available for a per-test fee at www.unl.edu/buros. You can also access the yearbook in print form at most college and university libraries.

Michael S. Matthews, PhD

Michael S. Matthews is a research fellow with the Duke University Talent Identification Program. He received his PhD in educational psychology, with a concentration in gifted and creative education, from the University of Georgia.