The Intelligence Quotient- or IQ- is one of the most popular subjects in psychology, often considered common knowledge. Despite IQ often being used as a shorthand for intelligence, even being used to define people, it seems like many people who use it don’t always know exactly how it works. IQ could be compared to the Chinese Room argument for AI- a person goes into the testing room, and comes out with a IQ score, so the score is assumed to be a function of their internal workings even if the process of how it got there isn’t clearly known.
IQ can be described as the BMI of the mind: a number gives some useful information for the typical mind/body, but still needs to be regarded with caution, especially in atypical minds/bodies.
For example, BMI is almost-useless in athletes, because the weight of muscles means that the majority of athletes will score as overweight or obese even though they are fitter than the average person. In the same way, while IQ is a useful estimate for a typical person in a familiar situation, it is less useful in people with neurodevelopmental disorders, or people from cultures that don’t have the same degree of standardised testing, to name just two examples.
Seeing as when it comes to IQ testing and how it works there is almost no such thing as freedom of information, I have no qualms about sharing a look into the methods and processes involved.
In a workshop session, we got to use the British Ability Scale testing kit, and in my dissertation experiment, I used the Weschler Adult Intelligence Scale-IV. (For participant confidentiality reasons etc, I couldn’t get any photos of the WAIS in use, unfortunately).
Even though the BAS is for children, and the WAIS for adults,many of their subtests are very similar. Childrens’ tests normally run along the same principles as adults, though with simplified questions and additional props such as small toys.
Assessments such as the BAS and WAIS aren’t just one IQ test with one direct result. Instead, there are multiple subtests measuring one area of ability. Subtest raw scores are compared to normed samples (tables of typical scores for people of various demographics) then summed to specific instructions. The test produces a set of sub-scores- the WAIS uses Verbal, Performance, and total IQ.
Full- Scale (total) IQ is usually an average of Verbal and Performance IQ, but it cannot always be calculated for a person. This mostly happens if there is a large discrepancy between Verbal and Performance IQ that would make the average too unreliable. For example, someone with a learning disability such as dyslexia may have a significantly higher Performance IQ than Verbal.
Generally, if scores between the two sections differ by more than one standard deviation (15/16 points depending on the test), that’s an indicator of some kind of learning disability or difference.
IQ tests have a lot of uses, especially in the USA, for diagnosing learning disabilities and other academic issues. However, some of their other uses can be a lot more controversial, especially as they are such a closed field.