The short answer to the question Are IQ tests an accurate measure of a person’s real intellicence would be no.
The long answer is that it depends both of the definition of intelligence, the definition of IQ and the psychometric properties of the test used. Unfortunatly I have no idea what kind of IQ tests you use in the US schools,
IQ, or intelligence quota, in the classical sense, is a very outdated concept as a measure of the construct we call “intelligence” in normal language.
The classical IQ tests are the Stanford-Binet test and the large amount of similar test. They measure some logic, analytic and verbal reasoning, as well as short term memory functions. To understand what these tests measure, we must take a look into the boring world of psychometric. Are you familiar with the terms stability, reliability and validity? If so, skip the next paragraph:
Stability is a measure of how stable the variable you want to measure is across time. In the beginning of the 20th century when the classical IQ tests were born, intelligence was supposed to be stable, inherent trait rather than something you can aquire. Thus, the classic IQ tests are supposed to show high stability, ie an individual should score about the same result in the test when taking it at different ages - simply put, your IQ should not change a lot with age.
Reliability is a measure of how accurately you measure the variable you want to measure. As you understand, high stability is a requirement for high reliability if you measure something you expect to be stable. If intelligence is stable, but a person scores IQ 98 at age 10, IQ 120 at age 20 and IQ 70 at age 40, then there’s either something wrong with the person, or the test is not accurate, or it lacks validity. There’s also something called test-retest reliability. If you give the same test to the same person let’s say 2 days in a row, the score should be about the same. If it’s not, something is also clearly wrong.
Validity is a measure of whether the test is measuring what you want it to measure, or something else. How to aquire validity is a long story, but one common way is to correlate your test to other measures you think are relevant to the variable you want to measure.
Classical IQ tests are a very good measurement of current level of education in the Western culture educational system, and it’s also a good predictor of aptitude for future studying in our educational system. Why is this? It’s because most of the classical IQ tests were validated against educational level to start with. We want to measure intelligence. Here’s a test. How do we know it measures intelligence? People who have a high educational level must be intelligent. Therefore, if highly educated people score high on the test, whereas people with low education score low, we are measuring intelligence. Thus, the majority of the IQ tests were validated.
The index “IQ” was obtained by giving the test to a lot of kids in different ages and calculating the mean for each age group. Thus you set norms to your test. 5-year olds scores X points, 10-year olds score Y points. IQ is the quota between the chronologic (real) age of the kid and the “intelligence age” or “mental age” as points in the test * 100. If a 5-year old scores Y points, ie the result that is usually scored by 10-year olds, the kids IQ would be 10/5 * 100 = 200. This is the reason why it’s children and young people can score extremely high on classical IQ tests, whereas a 60-year old can never score as high. Does this mean the 60-year old Nobel proze winners are not very intelligent?
Much has happened since the classical IQ tests. The Stanford-Binet has been revised many times, and corrections for error sources have been made. The Weschler scales were introduced in 1939 or thereabouts, and they use a different way of calculating intelligence, based on norm samples and standardized scoring. The WAIS and WISC score are not usually called IQ by professionals, but there is a way of transforming WAIS/WISC results to IQ although the test constructors strongly advice caution in doing so. IIRC the roof is set to IQ 140 in WAIS, since it’s not regarded meaningful to differentiate higher than that. Still, the Weschler scales also correlates highly with level of education, and it can’t be used for cross-cultural even between European countries or Europe/US. The US norms for instance, are significantly lower than the norm in my country. If the US norms was used here, a majority of people would score above the mean, and thus the mean would not be the mean. (Actually, the US norms are only used for forensic populations, ie criminals who have committed crimes with a minimum of 2 years prison as a penatly.) Does this mean my countrymen are more intelligent than Americans?
Here, classical IQ test are only used by the military (for recruitment) and by MENSA. Tests like WISC and NEPSY are used for children, mostly to identify special learning disabilities or as a part of a diagnostic procedure. For adults, “intelligence” tests are almost exclusively used for clinical assessment, as tools of diagnosis. But the tests are not called intelligence tests any longer, they are called neuropsychological or cognitive tests.
In research, the term “intelligence” is far to vague to be useful. In the 1960’s “intelligence” used to be defined as “what is measured with an IQ test” just as depression was once defined as “the condition that improves by giving TCA antidepressants”. Nowadays, the term cognitive functions is used when we refer to abilites like different memory functions, ability to concentrate, ability of learning, process-speed, logical reasoning etc. “Cognitive functions” is a much broader concept that what people in general mean by “intelligence”, and it includes a lot of functions that is not correlated to aptitude for academic studying.
The more modern concept G-factor is probably the closest you can get to what is commonly meant by “intelligence”. If you want to know more about the G-factor or other things related to “intelligence” ask, and I’ll try to answer your questions.
Btw, what IQ test are used in American schools?
