Make Your Child Smarter

True IQ

Q: How many IQ tests (WISC) are needed to estimate a "true" IQ. I have always heard three but can't find it in writing.

A: Neither can I! However, three tests would probably yield the best estimate, though still unlikely to be totally accurate. In fact, an estimate of a true IQ may not be revealed unless comparison is done between people of the same age, sex and relevant background. IQ tests are at best baseline estimates of one's true IQ. Perhaps, the fairest estimate is to take the sum of several scores on IQ tests and to divide them with the number of tests. In addition, the tests should be taken within a relatively short period of time between one another to be closest to accurate, and usually under one year for three or more tests.

A little more on IQ testing; a serious criticism of using only one test to assess IQ is the varying strength of people in different areas such as verbal skills, logical aptitude or spatial visualization than in others. Tests that capture certain skills that an individual is strong in would definitely produce better results than areas that the individual may be weak in. It is also found that to make an IQ score which is lower than one's true IQ is easier than to make it higher. For example, factors that may affect test results are taking tests that captures skills that one may be weak in, testing on a bad day (emotionally, physically, etc.), or even spending too much time on a few difficult items could indeed lower one's scores. Due to all these reasons, naturally, taking more than one test would produce better results.

Another important thing to note on true IQ is the standard error of measurement (SEM) of IQ tests which is the standard deviation times the square root of one minus the reliability. This sounds too technical – but here's an easy way to understand. If you were to take the same IQ test repeatedly over a certain period of time (with short intervals within a year, for example), with no change in the level of intelligence (you did nothing to “increase” your intelligence), it is possible that some of the scores would be higher or lower than the score that reflects your actual intelligence level (which is the true IQ score). The difference between your true score and your highest or lowest hypothetical score is the SEM. The bigger the SEM, the more the observed scores may vary from the true score and this means that the test is less reliable. In conclusion, the SEM actually tells us how accurate an estimate the IQ observed score is for what we think of as a true IQ score. This is exactly what all psychological test scores are – estimates only, and not absolute.

>> Back to Ask an Expert