What teachers need to know about intelligence – Part 1: Why IQ matters

Intelligence is required to be able to know that a man knows not.

Montaigne

Although it’s become a truism to say we know relatively little about how our brains work, we know a lot more now than we used to. Naturally, everything we know is contingent and subject to addition, but that doesn’t mean we can ignore it or pretend we don’t know enough to draw some fairly clear conclusions. Despite the many myths surrounding it, intelligence is a good candidate for being the most well researched and best understood characteristic of the human brain. It’s also probably the most stable construct in all psychology. So, what is it?

Intelligence has been described variously, as a capacity for logic, understanding, self-awareness, creativity, problem solving and the ability to learn new information more quickly. According to some it’s the ability to acquire and apply knowledge, while others see it as plain old ‘good sense’. Whatever it is, it seems safe to agree that it’s not simply one thing.

One generally accepted definition is that intelligence is

A very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience. It is not merely book learning, a narrow academic skill, or test-taking smarts. Rather, it reflects a broader and deeper capability for comprehending our surroundings—”catching on,” “making sense” of things, or “figuring out” what to do.

This being the case, who wouldn’t want to be cleverer? This is a far cry from the oft-repeated truism that all IQ tests show is how good you are at taking IQ tests. In actual fact, IQ has more predictive power than any other psychological construct. I’ve written before about the differences between fluid (Gf) and crystallized intelligence (Gc) and why we should probably concentrate on trying to raise crystallized intelligence and stop worrying about its fluid cousin. Here’s a brief summary of those differences:

An IQ test attempts to measure both Gf and Gc to provide an estimate of an individual’s general cognitive ability.

Before we go any further, we also need to understand that intelligence and IQ are not the same thing – IQ is just a proxy for measuring intelligence. That said, it’s a very robust proxy. Intelligence is measured by a battery of tests designed to assess a range of different mental capabilities. These typically consist of reasoning ability, memory, knowledge (including vocabulary and general knowledge) spatial ability and processing speed.

Here are a few examples of the sorts of questions asked in an IQ test. Matrix reasoning tests are designed to test fluid intelligence because it’s supposed to be possible to work out the answer without any prior knowledge. Here’s a fairly straightforward example:

Select a suitable figure from the four alternatives that would complete the figure matrix:

They can get a lot harder. An example of something that would test crystallised intelligence is a vocabulary test. Clearly, you can only answer the following question with any confidence if you have previously encountered the answer in some other context:

What is the best synonym for dismay?

  1. display
  2. jealousy
  3. provocation
  4. disappointment

A test of verbal reasoning requires you to engage in some reasoning for which you won’t be expected to have any specialised prior knowledge, but the better your vocabulary and general knowledge, the easier you’re likely to find it:

Afro-Eurasia is the largest landmass of all time.

  1. True
  2. False
  3. Cannot say

A properly conducted IQ test takes around 2 hours to sit.

Intuitively, we tend to believe that individuals might have a great memory, but struggle to add up columns of numbers, or possess extensive vocabulary knowledge but struggle to see patterns, or score well on tests of processing while being a bit rubbish at, say, rotating 3D images. It turns out that people who do well at one aspect on an intelligence test, tend to do well at all aspects. In other words, intelligence seems to be general.

In order to understand this counter intuitive finding we have to resist the temptation to think in anecdotes and instead attempt to think statistically. Claims, which may not be true for individuals, may well be the case when we make them about a representative sample of a population. The claims I’ll make about intelligence in this and subsequent chapters are true on average. They are probabilistic. Of course there will always be individuals who defy probability, but that doesn’t mean we’re unable to identify patterns that are the case for the majority. These patterns are correlations. That is, imperfect connections between two variables.

So, while I’m not claiming that a high IQ score causes people to be more creative, or to live longer, there is a correlation which is down to more than chance. This could mean that creativity causes intelligence, or that some other factor – maybe social background – cause both intelligence and creativity. Or it could be some other combination of all three. The point is, the connection is not random. When I say that people who are good at reasoning also have better vocabularies, I’m saying that this isn’t a matter of luck. So while correlation is not proof that one thing causes another, causation is implied. The same is true of the finding that when we look at IQ test results for a reasonably large sample of people, their scores in different areas correlate positively. This what is referred to as the general factor of intelligence. Or the g-factor. Or just g. Nobody knows for sure exactly why this should be the case, but nevertheless, intelligence researchers agree that it is.

Maybe the most surprising finding to come out of intelligence research is that intelligence is a good predictor of longevity; the higher your IQ, the longer you’re likely to live. It’s difficult to understand how the kinds of things measured by IQ tests could correlate with lifespan, but correlate they do. If we’re to understand why it’s so important to make kids cleverer we need to try to get to grips with this baffling discovery.

It seems much more reasonable for longevity to correlate with physical health, and indeed it does, but better health is also positively correlated with higher IQ. This might have something to do with the connection between intelligence and education (more on that later) as the better educated you are, the more likely you are to know how to take care of yourself and to afford to do so, because, you guessed it, intelligence also correlates with success in the workplace.

It’s probably not too much of a shock to find out that higher IQ scores are related to higher income, but more surprising is research that demonstrates the correlation with workplace performance. The more intelligent an employee, the more conscientious they’re likely to be. In fact, the correlations just keep coming as we can see:

There are pretty strong correlations with performing in jobs designated as ‘high complexity’ and ‘medium complexity’ and intelligence even makes a difference when the job is relatively straightforward. IQ even shows correlations with leadership and creativity indicating that, on average, cleverer people will have more ideas and be better at getting others to implement them.

It’s often argued that IQ tests don’t measure important cognitive qualities like creativity, but studies into the links between g and creativity have shown a moderate positive correlation. One problem is that creativity is itself tough to pin down so researchers rely on proxies like ‘divergent thinking,’ but people who do well on IQ tests also do well on the alternative use test “How many uses for a paper clip can you come up with?” (Nusbaum & Silvia 2011) More interestingly, they also do better on tests of musical discrimination (Mosing 2014) and have even been shown to register more patents for new inventions and receive more awards for artistic endeavours. (Wai, J. et al. 2005

If you’re not already convinced of the importance of intelligence, then what about the fact that higher IQ is positively correlated not only with better physical health, but also better mental health. The more intelligent you are, the less likely you will be hospitalised for a psychological condition. Contrary to many popular myths, there’s an especially strong link between higher IQ and lower likelihood of suffering with schizophrenia.

Whilst it seems to be the case that a higher IQ makes for a happier, healthier, longer life, clearly it can’t be the only causal factor. Before we get too excited, we need also to consider the connection between social class and intelligence. We also need to see whether we’re pointing the causal arrow the wrong way round. Maybe higher social classes, with its many attendant advantages causes, higher intelligence. This is certainly plausible. The correlation between social class and IQ scores is certainly there (r=0.3-0.5) but it’s not as strong as you might expect. There’s no doubt that your background has an effect on your life chances, but there’s also evidence of a connection between higher IQ and social mobility; the cleverer you are, the more likely you are to improve your socio-economic standing. Where you start off will be an important determining factor in where you end up, but so will your intelligence.

In addition, intelligence isn’t just an individual good; it’s also a social good. Raising IQ will also make society better for everyone. The more intelligent you are, the less likely you are to commit a violent crime and, even more interesting, the less likely you are to have a violent crime committed against you or of being murdered. In his magisterial study of violence and its decline in the modern era, The Better Angels of Our Nature, Steven Pinker argues that the cleverer you are, the more likely you are to cooperate with other people, weigh the consequences of your actions and think in such a way that allows you to escape the confines of your own limited experiences. Amazingly, the more intelligent you are the lower the probability that you’ll be racist or sexist and the greater the likelihood that you’ll be socially liberal. And if all that wasn’t enough, there’s also a pretty strong connection to happiness. People who score more highly on IQ tests also tend to report being happier.

Possibly the strongest correlations with intelligence is that with educational outcomes. The higher your IQ, the more likely you are to do well in school and the longer you’re likely to stay in school. It might seem self-evident that people who score better on IQ test go on to do better a school, because educational success is predicated on measuring the same kind of ability. But, it’s worth knowing that the correlation between IQ tests taken five years previously and GCSE results taken at age 16 is one of the strongest yet found in psychology (r=0.81). The predictive power of the IQ test seems astonishingly accurate.

Accurate, but far from perfect. Before you start wondering whether we could do away with examinations in school altogether and just give a single, one hour IQ test, you need to know that some people with high intelligence did not perform well in other tests and that some people who did well in their GCSEs did poorly on an IQ test. There are, it would appear, other factors such as conscientiousness, motivation and self-control, involved in educational success. There’s also evidence that education may cause increases in intelligence.

Whatever the question is, it seems that intelligence is always at least part of the answer. If we want children to be happier, healthier, more successful, to live long and earn more, then it will pay to try to make them cleverer. So, having tried to established that IQ matters, in Part 2 of this series I’ll discuss the fact that IQ scores seem to be rising across the developed and developing world and what this suggests about intelligence, IQ testing and improving education.

Acknowledgement: I owe a huge debt to Stuart Ritchie’s wonderful little book Intelligence: All That Matters for collecting together most of the research that has informed this post.

26 Responses to What teachers need to know about intelligence – Part 1: Why IQ matters

  1. Michael pye says:

    Clearly reasoned. Enjoyed reading it.

  2. Robert A. Gordon says:

    At last, a well-informed and responsible discussion of human intelligence on the internet! Most internet discussions seem to regurgitate Howard Gardner’s notion of multiple intelligences as though it superseded the science of general intelligence or “g.” It does not come close. Gardner’s views have become fashionable among educators and parents, because he provides more opportunities for conveying “good news” for them and their constituents when IQ scores are not as high as some might prefer. Yet, as this blog points out, it is IQ (a good measure of g) that does most of the cognitive heavy lifting in life. In principle, there is no limit to calling performances of any type an “intelligence,” except the gullibility of the audience. “Crocheting intelligence,” for example, might have special appeal to persons skilled in that craft, when terms such as “expertise” and “knowledge” would serve just fine without undermining a major psychological construct that continues to affect outcomes whether we acknowledge it or not. Do your friends and the world a favor by forwarding this entry to them. It is especially valuable for highlighting the importance of the intelligence context that every person finds themselves in. This context is an often overlooked resource for explaining good individual decisions. Consider: Would you rather consult for advice someone who is smarter than you, or not as smart? The differential availability of such persons in our individual social lives matters to all of us. Thus, we all have a stake in one another’s intelligence.

    I strongly recommend the January-February, 1997 special issue of the journal Intelligence, vol. 24, No. 1, guest-edited by Linda S. Gottfredson, as a source of authoritative overviews and new developments consistent with the science of g. It contains a list of 25 propositions often contested in popular media that are actually mainstream, backed by 53 signatories who are distinguished researchers. Three of its articles won Mensa awards for being among the best in their year of publication. It is still a good place to begin for those interested in understanding the most important of all human cognitive abilities. One should keep in mind while reading it that “mainstream” does not exclude cutting edge research from worthiness. It simply means that the 25 propositions are far less controversial among experts than many journalists and academics would have the world believe.

  3. Andrew Darling says:

    Hi. Fascinating as ever. One thing I’m confused by is your statement that fluid intelligence is fixed but decreases after your late 20s…

    • David Didau says:

      Sorry, that was a bit over-simplified. Basically Gf increases with age in line with executive function but, as yet, no one has found a way to increase it beyond this. Gf is very closely correlated with working memory capacity which is equally resistant to attempts to increase it. Does that help? If you need more, try this: http://www.learningspy.co.uk/psychology/making-kids-cleverer-part-2/

      • Andrew Darling says:

        Thanks – yes, that does help. I’m applying this to Year Ones. Your writing has really helped. I’ve realised that my SEND children can’t blend CVC words because they can’t hold the 3 sounds in their heads. So we’ve introduced onset and rhyme – the children learn ‘-at’ as a unit of sound – push it into their long term memory – and then blend it with s, b, p, etc.. the results have been spectacular in just the last 3 or 4 weeks since we started. If fluid memory increases with age then that would help to explain why this age of children can’t follow 2 step instructions and so on. Thanks again.

    • Robert A. Gordon says:

      If I may attempt added clarification of the description of fluid intelligence as “fixed,” and the superficial paradox that it is also described as “changing”: In this context, fixed means not subject to environmental influence as far as we can tell. Thus, although fluid intelligence increases during early maturation, and starts decreasing in the late twenties, it goes its own merry way, regardless of normal environmental efforts to change it either way (i.e., excluding brain-damaging hammer blows to the head, etc.). It seems to parallel age-related changes in brain size, which in the late twenties reflect losses in brain cells. Crystallized intelligence, on the other hand, can reflect environmental inputs until very late in life, so that if one learns arithmetic well in school, one will still know arithmetic in old age if one is healthy, but if one is not taught arithmetic, even though one was smart enough to learn it, one will not know arithmetic later in life (hence the crystal metaphor to contrast with the fluid metaphor). Thus, accumulated learning, such as arithmetic, is available to apply to suitable problems later on, and represents a gain or a resource. Jensen suggested for crystallized the description, “consolidated knowledge.” Fluid intelligence seems less dependent on formal learning and so is what seems to be at work in attacking truly novel unsolved problems. But when applied to learning, it seems to be responsible for producing consolidated knowledge and hence crystallized intelligence. Among physicists and mathematicians, there has been a concern that if one has not done great creative/theoretical work by one’s late twenties or so, one is unlikely ever to do so. And even if one has done such work, one is unlikely to do so again late in life. In those fields, creative work is usually extremely novel. But in, say, history or the social sciences, where great work may depend on a lifetime’s accumulation of information and perspective, one’s best work is more apt to unfold late in life as a synthesis of what has been learned previously, and hence it depends on that learning heavily. Einstein’s “miraculous year” occurred in his mid-twenties, and his General Theory was developed in his mid-thirties. He had an unusually long creative period, extending even to his forties. But when working on his General Theory he had to turn to his friend Michael Besso, a mathematician, for tutoring in non-Euclidean geometry, thus repairing a gap in his crystallized intelligence that was crucial for solving the problem in hand. The two forms of intelligence are highly correlated with each other, of course, and both can figure to different degrees in solving problems, according to the nature of the problem and the background of the individual.

      • David Didau says:

        Thank you – very helpful

      • Robert A. Gordon says:

        I did not mean to suggest chunking had no limit, only that it increased the number of original elements that could be easily retained. The old telephone numbering system in the US, prior to ten-digit Area Codes, employed verbal “exchange” names to reduce the number burden on WM. Thus, the entirely numerical phone number 962-6114 would be treated as WOrth 2-6114. No one remembered that WO fell on the numbers 9 and 6 on the dial. Certainly, that system was easier for me as a child, and the exchange names applied to many phones in an area and hence became familiar in and of themselves.

  4. Pique Boo says:

    “IQ scores seem to be rising”

    Not necessarily. I recall Flynn saying via some research in the last year or two that the (coined by Murray) ‘Flynn Effect’ had reversed in the UK. I have seen the same noted for a few other countries. Then we have whether IQ score rising means anything. Is underlying ‘g’ rising or is [whatever] just making people better at IQ tests, and if so what is the impact on anything else?

    This, the Yulia Kovas key-note from ISIR 2016, has a strong education angle and is worth the time of day (55 mins): https://www.youtube.com/watch?v=ro-upowRUzU

  5. T. M. says:

    Really enjoyed this. Very thoughtful. I think the likes of yourself and Daisy Christodolou are doing some important work in raising awareness about the importance of imparting knowledge. However, the point you make about the correlation between intelligence and job/ economic success seems a bit odd on an education blog. Surely you are imploring teachers to raise the intelligence of all pupils. If this were succesfully carried out it could not affect individual social mobility, as the individual student who’s intelligence had been raised would find herself competing in the job market against other candidates with raised intelligence.

    • David Didau says:

      Ha. I guess I see intelligence as more of a social good than an individual good. Social mobility is a double edged, zero-sum sword. If one goes up another has to come down. But, a commitment to raising the IQ of all students means that this will be less about SES.

  6. doylea36 says:

    Are current IQ tests too culturally contingent on the current idea of intelligence?

    Is the reason a person with a high IQ lives longer is that they came from an environment that provided more practice with the things tested and are therefore more likely to be middle class and have better healthcare, better food, better living conditions and a better education.

    How do we compare differences in the results between ethnic groups, nationalities or cultures?

    • David Didau says:

      1. Maybe, but maybe this isn’t a bad thing. There’s a good argument that the cultural biases reflected in IQ tests help us make more moral choices.
      2. No, the effects of IQ on longevity have been controlled for SES.
      3. That’s a can of worms. My question would, why do you want to?

  7. Sergej Visser says:

    Looking forward to pt 2.

  8. […] Part 1 of this series I laid out why IQ matters and that, far from being a banal measure of merely of how […]

  9. […] I explained here, the scientific consensus is that intelligence is general. That is, if you good at verbal […]

  10. […] make children more creative and better critical thinkers, we need first to make them cleverer. See here for my views on IQ and […]

  11. […] Didau has (obviously) written about this and his blog is as usual essential reading. The second part of that blog puts forward the idea […]

  12. […] I discussed here, there’s a considerable body of evidence that IQ scores have quite impressive predictive […]

Constructive feedback is always appreciated

%d bloggers like this: