Myth 15: IQ Tests Are Biased against Certain Groups of People (50 Great Myths of Popular Psychology)

Regarding IQ tests, there are three important questions that are asked. 1) Do IQ tests predict success? 2) Are IQ tests indicators of intelligence? 3) Are IQ tests biased towards certain groups of people?

The first question is debatable. Some argue that IQ tests can only predict one thing – how well people do on IQ tests. Others argue that IQ tests can reliably predict which students will become more successful in their careers, and even which presidents will accomplish more.

As for question 2, intelligence is defined as the broad capacity to understand things and solve problems but cannot be constricted to some narrow academic skill or book smarts.

 And for the third question, there are many different viewpoints. Some people, such as Ralph Nader, have called for the abolishment of the SAT’s because minority groups do not do as well on them. And in the U.S, there is a good reason to believe that racial prejudice exists.

In the influential Larry P. v. Riles (1980), the 9th District court of appeals in California ruled that an unbiased test should yield the same pattern of scores when given to different groups of people, and placed strict limits on the use of intelligence tests for classifying children as mildly mentally retarded for educational purposes. Another court case saw the Golden Rule Insurance Company sue the state licensing board and test publisher because less black respondents (than white respondents) answered correctly on some items on the licensing tests.

However, just because two groups of people may differ in some ability, this does not indicate that the tests are biased, but may indicate that their abilities are biased. If a doctor says that his measurements tell him that men are taller than women, it does not mean that his tests are biased, but that men are, on average, taller than women.

Some psychologists think that there is truth to the claims of test bias. Here is why: the items making up the test, and not the test itself, can be biased. This phenomenon is called differential item functioning or DIF. For any pair of groups (men versus women, or blacks versus whites), we can examine each item on an IQ test for DIG. If members of two groups perform the same on the test except for a particular item, we can see this as evidence of item bias.

Researchers commonly discover that many IQ test items meet criteria for DIF. But how can themselves demonstrate DIF without scores on the entire test being biased? It turns out that many of the DIFs are trivial in size. Further, some items favor one group, while other items favor another group, so their effects cancel each other out. DIF does not necessarily produce test bias.

The verdict according to the authors is that IQ tests validly predict performance in many important areas in life, with no evidence of bias against women or minorities. The real bias is when the blame is cast on the IQ tests themselves, without considering environmental explanations, such as cultural disadvantages, for differences in test scores across groups.  

"Silence is the best expression of scorn" - G.B. Shaw

This site uses Akismet to reduce spam. Learn how your comment data is processed.