By Liam Mannix
I have a friend – a very smart one – who refuses to sit an IQ test. She won’t even do one of those hokey ones on the internet, or the quick ones that used to pop up on Facebook.
“I’m scared of getting it tested,” she tells me. “What if it’s a lower score than I expect?”
Indeed. Since Intelligenzquotient was coined by the German psychologist William Stern in 1912, this single metric has come to dominate the way we think about intelligence.
IQ tests are powerful - but are they useful?Credit: Alamy
Given how complex intelligence is, that mere fact is remarkable. But IQ itself is a remarkable measure. Evidence suggests it is very powerful – and predictive. And like most powerful things, it can, and has been, co-opted for both good and evil.
In today’s Examine, we present two views on IQ – its potential, and its potential pitfalls.
‘A robust, predictive measure…’
At the core of IQ is a key insight: people who do well on one test of intelligence tend to do well on others. This is actually quite surprising, if you think about it. Why should a person who is good at maths also have a large vocabulary?
The answer is general intelligence – our ability to reason, plan, problem solve, think in the abstract, understand complexity and learn quickly. People who are good at one tend to be good at the others.
There are a lot of different tests that scientists use to measure IQ. But the key point is these tests are relative, they will have been given to thousands of people.
This allows researchers to measure IQ across a population and then fit that score onto a normal distribution – a bell curve – so that the average score is 100.
This means two thirds of people in a given population will have an IQ score between 85 and 115.
IQ tests are standardised to a bell curveCredit: DMCQ / Wikipedia
Other cool IQ things:
- Heritability estimates – how much your IQ comes from your genes, rather than your environment – sit around 50 per cent.
- That means IQ can be changed by manipulating a person’s environment, especially in children. Better nutrition and more education have been shown to be able to raise IQ.
- Intelligence does change over time, but not by a lot after you’ve reached your teenage years. One study tracking people from age 11 to 90 found intelligence at age 11 strongly predicted intelligence at age 90.
- One finding I find striking: intelligence inheritance changes as a child ages. In infancy, about 20 per cent of intelligence is inherited, but that rises to 60 per cent by adulthood. Why? Perhaps more genes activate as we develop? Or perhaps children who are deemed more gifted are given greater access to education as they age?
Does IQ matter?
You can create all sorts of metrics to measure people – the number of bumps they have on their head, for example. What makes IQ unique, important and useful, says Curtin University cognition researcher Dr Frank Baughman, is it “is a very powerful predictor for a range of life outcomes”.
Meta-reviews of IQ studies generally show IQ partially predicts a person’s salary. There is also a correlation between IQ and educational outcomes and occupation, although factors such as your parents’ IQ, occupation and socio-economic status also play a part.
IQ even seems to predict mental and physical health and how long a person lives.
Assortative mating – the phenomenon in which people tend to choose partners with similar traits – is stronger for intelligence, particularly verbal intelligence, than almost any other trait.
“It’s a very good measure,” says Baughman. “It’s absolutely precise in its ability to put one person’s score relative to a population.”
“It tells you very little about a person”
But there are several reasons to be sceptical about the true usefulness of IQ.
First, putting metrics on humans has not gone well, historically. IQ tests were developed with the noble aim of finding kids in need of extra help at school, but were quickly co-opted by eugenicists to argue for sterilisation.
And as we’ve seen, intelligence is at least partially determined by a person’s environment. If people have more access to opportunity, they are more likely to record higher IQ scores.
“They’re going to miss people who haven’t had opportunities to access education, haven’t had books in the family home, aren’t familiar with test taking,” says Dr Michelle Ronksley-Pavia, a gifted-education researcher at Griffith University.
If you sort people based on IQ, you’re also sorting them based on the opportunities they have had in life – and simply giving a high-IQ child access to more schooling might reinforce the cycle of privilege. Because of this, Ronksley-Pavia says the gifted-education field now uses far more comprehensive tests to identify giftedness, such as the Raven’s Progressive Matrices test, which measures abstract reasoning.
IQ testing focuses on certain domains such as visual, mathematical and language abilities but ignores other important ones, such as creativity or divergent thinking.
Then there’s the broader point: how useful is it, really, to be able to predict someone’s future income or job status? Particularly when those outcomes are also influenced by their socio-economic background.
“An IQ score,” says Ronksley-Pavia, “tells you very little about a person and what they can do.”
Liam Mannix’s Examine newsletter explains and analyses science with a rigorous focus on the evidence. Sign up to get it each week.