IQ Testing, Cultural Bias, and the Limits of Intelligence

What Is IQ?

According to Holden (2003), IQ tests measure individuals for "g," or a general intelligence factor that is supposedly inborn. A debate has been ongoing among psychologists — the originators of IQ testing — about the importance of the IQ score. "IQ scores predict a wider range of important social outcomes than perhaps any other psychological trait… bright people have a tail wind in virtually all aspects of life" (Gottfredson, cited in Holden, 2003). According to Gottfredson, IQ affects nearly every aspect of an individual's life from birth to death. As one example, she describes the healthcare people receive. If a person cannot read and understand the directions on a medicine bottle, the level of care they receive is likely to be diminished. If they cannot understand a basic consent form, they may agree to surgeries and treatments that a more informed patient would decline. This researcher correlates low IQ with "higher health costs, poorer health, and more frequent hospitalization" (p. 193) and claims that people with low IQ scores have more accidents as well.

What she does not acknowledge, however, is that an inability to read well does not necessarily mean a person is not intelligent. Consider the case of a man whose alcoholic mother neglected him severely and did not send him to school until he was ten years old. When he registered, the school assumed — based on his age — that he should be placed in fourth grade, never realizing he had missed learning to read entirely. He did not learn to read until he was forty, when his own daughter taught him. IQ tests taken during his school years would almost certainly have recorded a low score, yet labeling him as low-intelligence would be both unfair and inaccurate. He is, in fact, a very intelligent man.

In another study, Hendrix (1998, cited in Holden, 2003) used data from a longitudinal study — the 1998 National Longitudinal Study of Youth — and found that some children were "resilient," showing no signs of "depression, delinquency, or school failure" (p. 193). The most resilient children had the highest IQs, while the least resilient had the lowest. Deary (cited in Holden, 2003) reached similar conclusions about older adults: those with higher IQs enjoyed a better quality of life, lived longer, were healthier, and functioned more effectively in daily life.

Multiple Intelligences

Howard Gardner's work, by contrast, views intelligence far more broadly than an IQ test score can capture. His theory of multiple intelligences is grounded in what he calls "brain evidence for its existence in the nervous system" (Gardner, cited in Hogan, 2002). This implies that one or more combinations of intelligences evolve within individuals in response to experiences and challenges accumulated over thousands of years. According to Gardner, all individuals possess at least eight intelligences — linguistic, logical-mathematical, spatial, bodily-kinesthetic, musical, interpersonal, intrapersonal, and naturalist — though some may remain underdeveloped. These intelligences tend to develop unevenly; a person may have strong linguistic ability and struggle with mathematics, for example.

Consider naturalist intelligence. A man who grew up on a farm and spent his whole life outdoors was skilled in the woods, knew all the wild animals and their habits, could describe a location by the curve of the land, its smells, sounds, and vegetation, and could predict a hard winter by observing the behavior and condition of animals. If such intelligence were formally tested, he would score exceptionally high — yet standard IQ tests do not touch this domain at all. IQ tests are limited to linguistic and logical-mathematical skills. Someone like that, who did not speak standard English, read little, and had no exposure to city or academic life, might easily be labeled "low IQ." The same can be said of Native Americans and Australian Aborigines, both of whom have historically scored low on IQ tests.

IQ and Cultural Bias

A central problem with IQ testing is that the people who design the tests are generally familiar only with their own culture — its assumptions, customs, and habits. The tests reflect that culture rather than the culture of the person taking the test. Michael Cole (cited in Nunley, 1995) conducted a cross-cultural study on children to determine at what age they learn "conservation" — the understanding that when water is poured from a short glass into a tall glass, or when a ball of clay is rolled into a long string, the quantity remains the same. Cole compared children from a West African fishing village to children from a Mexican community where pottery was the dominant occupation. When water was used for the test, the children of fishermen in West Africa had mastered the concept of conservation at an earlier age than the children of potters. But when clay was used, the children of potters had learned conservation at a younger age than the children of fishermen.

In another study, De Lacey in Australia (cited in Nunley, 1995) placed objects on a grid and asked children to remember their positions. White city children quickly memorized the placement of mass-produced objects such as scissors, safety pins, and nails, outscoring Aboriginal children who had no experience with such items. However, when the objects to be memorized were stones of different shapes, sizes, and colors and other natural objects, the Aboriginal children scored higher than the white city children. Both studies clearly illustrate what is meant by cultural bias: even a task as seemingly neutral as identifying patterns in a number sequence is not culture-free. Familiarity with the test situation itself can significantly influence results.

Even the way scores are computed has become a subject of research. Coyle (2001) conducted a study in which "worst performance" proved to be a better predictor of IQ than "best performance." The researcher re-examined data from earlier research in which elementary school children were tested for memory using the Wechsler Intelligence Scale for Children, Third Revision (WISC-III) to obtain IQ scores. Children were then shown words on index cards that could be sorted into categories — for example, boat, car, bus → TRANSPORTATION; apple, orange, banana → FRUIT; cannon, gun, sword → WEAPONS. They were given time to study the words but were not told that the words could be categorized or given any memory strategies. After ten minutes of study, the words were covered, the children completed an unrelated problem-solving test, and were then asked to recall as many words as possible from the first task.

The researchers found that a student's minimum performance correlated more closely with their IQ score than any other single variable. When predictions based on worst performance (minimum recall) were compared with predictions based on best performance (maximum recall), the worst-performance predictions were more accurate — indicating that "worst performance reveals more about intelligence than best performance" (p. 9). The study was significant because it measured preparatory strategies, something earlier research had not examined, and it was the first time the worst-performance rule was tested on children rather than adults. The findings suggested "developmental invariance" — that is, no meaningful difference between people of different ages — though the researcher notes this should be tested in a study that gives adults and children the same task simultaneously. Importantly, low-IQ participants sometimes performed well overall, but their performance dipped far lower than that of high-IQ subjects, making minimum scores particularly revealing.

Conclusion: Beyond the IQ Score

Intelligence is more than an IQ score. As Gardner suggests, there are different kinds of intelligence, and IQ tests address only two of them. When Georgia's public schools implemented a new system for identifying gifted and talented students — one which included but was not limited to IQ scores — they discovered significant intelligence and talent among minority students who had previously been overlooked (Krisel & Cowan, 1997). As a result of a research project conducted by the National Research Center on Gifted & Talented, the State of Georgia passed a law requiring public schools to use multiple criteria when assessing students for giftedness, rather than relying solely on the IQ score. The new framework retains IQ and achievement tests but adds other categories of intelligence, some of which require professional judgment rather than a standardized score.

IQ appears to be an important predictor of success, but it is not the only one. By expanding the identification process to include multiple criteria, the problem of cultural bias in IQ testing can be controlled, at least to a meaningful extent. A child with musical intelligence, for example, can be easily identified, and that talent will be evident regardless of IQ score. IQ matters — but it is not everything.

References

Coyle, T. R. (2001). IQ is related to the worst performance rule in a memory task involving children. Ablex Publishing Corporation, 29(2), 117. Retrieved 3 December 2008 from Expanded Academic ASAP database via Gale.

Hogan, R. (2002). Might outdoor education help develop intelligence? Outdoor Council of Australia, 6(2), 2. Retrieved 3 December 2008 from Expanded Academic ASAP database via Gale.

Holden, C. (2003). The practical benefits of general intelligence. American Association for the Advancement of Science, 299(5604), 192. Retrieved 3 December 2008 from Expanded Academic ASAP database via Gale.

Nunley, M. (1995). "The Bell Curve": Too smooth to be true. 39(1), 74. Retrieved 3 December 2008 from General OneFile database via Gale.

Shipman, P. (1994). Legacy of racism. National Review, 46(D), 5. Retrieved 3 December 2008 from Firstsearch database.