When you’re trying to have an impact, it’s useful to know how successful you’ll be in different careers so you can pick the right one. But how can you do this? There are a few predictors of success that have been studied by psychologists, but the results aren’t widely known. The scientific consensus is that the best way to predict someone’s career success is to assess their general mental ability (GMA), which is similar to what most people mean by “intelligence”. You might find this surprising, so I’m going to summarise the evidence backing it up. Then I’ll talk about:

  • Why GMA is so important in work – mainly because people with higher GMA learn faster.
  • Which other factors affect success – job complexity, personality, and experience.
  • What this all means for your career – choose jobs that fit your GMA and find the best ways to increase your chances of success.

What is general mental ability (GMA)?

In 1904, psychologist and statistician Charles Spearman noticed that children tend to get similar marks in unrelated school subjects and he thought that there might be an underlying factor affecting their performance. He developed the statistical method of factor analysis to find how many underlying factors there were.1 Since then, many psychologists psychologists have used factor analysis to conclude that there is a single underlying general (or g) factor that explains people’s performance on different cognitive tests, although there is some disagreement about this.2 No single test measures general mental ability, but it can be measured from the results of tests of more specific cognitive abilities. But what is it? One description is “the ability to deal with cognitive complexity.” “[M]ore complex tasks require more mental manipulation, and this manipulation of information – discerning similarities and inconsistencies, drawing inferences, grasping new concepts and so on – constitutes intelligence in action.”3

General mental ability predicts career success: the evidence

Systematic analyses of thousands of studies support the idea that GMA is a good predictor of career success. Depending type of job and how performance is measured GMA explains between 30% and 70% of the variation in people’s work performance (i.e. correlations of between .56 and .84), which is larger than any other known predictor.4

Before we look at this evidence in more detail, we need to know what we mean by success. There are 3 main measures of work success:
* Evaluation of performance on tasks similar to those encountered on the job (work-sample tests)
* Performance ratings by supervisors
* Position in the occupational hierarchy

All three are predicted by GMA.

First, GMA predicts performance on work-sample tasks and ratings of supervisors. Evidence from several meta-studies shows that when performance is measured using work-sample tests, the correlation between GMA and performance is 0.84. When supervisor ratings are used, the correlation is lower, at 0.74 for high-complexity jobs.5

GMA also predicts how high up you get in the job hierarchy – i.e. your occupational level.6 US Employment Service data shows a strong correlation (0.72) between GMA and occupational level and US military data shows that mean GMA scores are higher at higher occupational levels. Also, there is a wider variety of GMA scores at lower occupational levels than at higher ones. It seems that there are high-scoring people in low-level occupations, but low-scoring people are unlikely to get promoted to higher levels.7

But to fully show the link we need to track people with known GMA over time to see if high GMA individuals end up being more successful. This has been done.8 In a longitudinal study of 3,887 young adults, GMA predicted movement in the job hierarchy 5 years later. Another study found that if people were in a job that was less complex than their GMA would predict, they moved up to a more complex job and vice versa. The predictivity of GMA even holds when controlling for socioeconomic status by comparing biological siblings. “When the siblings were in their late 20s (in 1993), a person with average GMA was earning on average almost $18,000 less per year than his brighter sibling who had an IQ of 120 or higher and was earning more than $9,000 more than his duller sibling who had an IQ of less than 80.”9

The link has also been confirmed by two natural experiments.10 In 1978 a US Steel plant in Pennsylvania changed the way they chose people for their skilled trades apprentice programs. Before the change they filled their positions with people who got the highest results on a GMA test. The plant replaced this system with a test that most people passed followed by allocation of jobs by seniority. The plant’s performance records show that after the new system was introduced trainees learned less and dropped out more. And because apprentices took longer to get through the course, training got more expensive.

The second natural experiment happened in the 1980s in Washington DC. The police force changed their selection procedures for new applicants from a GMA to a non-GMA based selection. After the change, it became more difficult to train new officers. 80% of new hires couldn’t complete the required training, so the content of the training had to be made easier. But the police officers produced from this training were often incompetent, which caused lots of problems. For example, the solution rate for murder cases declined from one of the highest to one of the lowest in the US.

Why does general mental ability predict success?

The evidence for the link between GMA and performance is strong, but it’s made more plausible by if we can understand why GMA predicts performance. GMA mainly influences performance through the rate at which people learn knowledge relevant to the job – people with higher GMA learn faster.11 But GMA predicts success even when you take account of job knowledge. With high GMA, people are more able to go beyond existing job knowledge and make judgements in unfamiliar situations.12

Other relevant factors

Job complexity

Although GMA predicts performance in all jobs the more complex the job is13, the stronger the relationship between GMA and performance.14 And the more complex the job, the more variation there is between top performers and bottom performers.15 So if you have one of the highest levels of GMA in a highly complex job, you’ll have a high output compared to the average performer.

Personality

You might think that personality matters more than GMA – some people seem to have the kind of drive that leads to success. And this is true – personality does matter, but less than general mental ability.16 The consensus model of personality suggests that there are five fairly independent dimensions of personality. These are conscientiousness, emotional stability, agreeableness, extraversion, and openness to experience. Of these, the most valid predictor of success is conscientiousness with a correlation between conscientiousness and job performance of 0.31.17 So personality is predictive of success, but not as much as general mental ability.

One possibility is that success is related to personality characteristics multiplied by GMA. So maybe someone with low conscientiousness wouldn’t be successful, however high their GMA was. One study suggests that this may be true: “the combination of highly general cognitive ability and motivation is significantly associated with more early career success.”18

Experience

The importance of experience is unclear. Some studies suggest that experience is less important than mental ability.19 Even more surprisingly, the more experience you have, the worse experience is as a predictor of success.20 For people with 0-3 years experience, the correlation between experience and performance is .49 but this drops to .15 for people with 12+ years of experience. Conversely, the strength of GMA as a predictor increases the more experience you have. With 0-3 years of experience, GMA correlates with performance at .35 but this rises to .59 for people with 12+ years of experience. One source suggests that after a few years, additional experience doesn’t lead to more job knowledge and therefore doesn’t increase performance.21

However, this is highly unintuitive – some people spend decades developing expertise and don’t reach a plateau after 5 years. Top performers in most fields tend to be middle-aged (think of scientists, CEOs, chess-players). If experience didn’t matter that much after about 5 years then we’d expect to see a wider spread of ages at the top.22 Since fluid intelligence (the ability to reason using novel information as opposed to existing knowledge) declines as people get older,23 we’d expect the top performers in many fields to be young.

We can try and resolve this apparent contradiction as follows. Perhaps the research showing that the importance of experience declines is correct for many jobs because these jobs have a relatively small amount of knowledge that needs to be learnt. Once this is learnt, the main determinant of success is GMA because it predicts people’s ability to reason in novel situations. However, some jobs have much more knowledge and experience required, so it takes decades to become an expert.

Even further than this, we could divide jobs into those that require a great deal of knowledge and those that don’t require so much knowledge but do require a strong ability to reason in novel ways. Then we’d expect top performers in the knowledge-dependent jobs to be older and those in the novel reasoning jobs to be younger. This is what we observe:24 people in pure maths, theoretical physics, and lyric poetry tend to peak in their 20s or 30s, whereas people other fields such as history, novel writing, philosophy, and medicine tend to peak in their 40s or 50s.

Common objections to this research

I have covered the importance of personality and experience above, but there are other possible objections:

Aren’t specific aptitudes more important?

You might think that specific aptitudes (such as verbal, spatial, or numerical ability) would predict job performance better than GMA, since different jobs seem to need different specific aptitudes. Because of this, you’d expect that you could predict success better if you gave people specific aptitude tests and put more emphasis on the aptitudes that were most relevant to the job. It’s not clear whether this is true. One review paper25 suggests that if you make a test that weights certain specific aptitudes more than others, it doesn’t have much more predictive power than a general mental ability test. This paper suggests that specific aptitude tests are predictive, but they are predictive because they measure g as well as measuring the specific aptitude. However, there is a theory of cognitive abilities that says that there are approximately 8 broad abilities and there is some debate as to whether they can be summarised into a single general factor.26 Even more importantly, the correlations between these specific abilities is lower for people of high ability.27 So if you’re intelligent you should consider whether you have one type of ability which you are much stronger in than others.

Doesn’t GMA only matter up to a certain level?

It’s possible that for each job there is a certain level of general mental ability required, and after that level is reached there is not much benefit from being more able. If this was true, we’d expect to see a non-linear relationship between GMA and performance. But overall,
the relationship between GMA and performance seems to be linear. One meta-study28 analysed 174 studies involving a total of 36,614 people and tested for linearity. They found that nonlinear relationships were not found at levels greater than expected by chance. However, this study did not cover the highest and lowest ends of the occupational spectrum very well.29 Similarly, most of the studies I referenced earlier have been done on large populations of people with fairly normal jobs and so unusual patterns at the highest levels could have been missed. It’s still possible that the very highest jobs and achievements need only a certain level of GMA, and other factors are more important beyond that level. Malcolm Gladwell claims this in his book Outliers.30 He suggests that while Nobel prize winners are intelligent, they aren’t necessarily of the very highest intelligence. Gladwell’s evidence is quite weak – he supports his claims with single studies, anecdote, and quotes from experts. This kind of evidence is not strong in psychology, where results often turn out to not be repeatable. Overall, I haven’t seen strong evidence either way on whether GMA is predictive at the very highest levels. However, given that GMA is so strongly predictive for most jobs, I suspect that the pattern will hold even at high levels.

If GMA predicts success, why is it not more commonly known?

Many companies do not use tests of GMA in their hiring. If GMA is so predictive of work success, why would this be? Here are some possibilities, although I don’t have evidence on this:
* It may simply be that hiring managers don’t know about the evidence. The first meta-studies on the topic were done in the 1980s and so perhaps this knowledge hasn’t filtered through to hiring managers yet.
* It may be that employers are worried about legal or worker morale issues from using ability tests in job interviews.
* Some of the conclusions of this research are unintuitive – e.g. that experience often doesn’t matter as much as GMA for job performance.
* Perhaps they do select for GMA, but indirectly, through interviews and assessments of past performance.

What this means for your career

Some of your career options will be more of a gamble than others. Most accountants, for example, will earn a good salary and be able to donate a lot of money to charity. But in some careers only a few people can have an impact and they have a disproportionate amount of impact. Research is one example – only a few researchers make the most important breakthroughs. Entrepreneurship is another example – most new businesses fail but a few become so successful that their owners can donate hundreds of millions to charity. Because there’s more variation between people’s performance in highly complex jobs31 and because these jobs require higher GMA we’d expect that people with higher GMA will have a better chance of succeeding at highly variable careers like research and entrepreneurship. People with lower GMA might want to go into lower-variance jobs with more sure outcomes.

From personal experience, I’ve also found it useful to know that job knowledge is a strong predictor of performance. Since reading the evidence about GMA and success, I have concentrated more on effective learning and choosing the most useful job-related knowledge to learn.

Does this mean that you can’t do much to improve your chances of success?

On first reading this, you might be worried that there is very little you can do about your chances of success. But these results shouldn’t cause you to give up:

First, it may be possible to increase your GMA. I haven’t seen evidence on whether or not this is possible, but it’s worth investigating.

Second, impact isn’t the same as what most people call success. You could have far more impact than someone with much higher GMA just by taking opportunities for impact that other people don’t know about.

Third, the studies I’ve quoted didn’t look at whether some people achieved more success than you’d expect from their GMA. Perhaps some people used better methods of working. We know from the GMA and success research that job knowledge is an important predictor of success. So how can you learn more effectively? This paper32 (summarised here) shows that the methods most commonly used by students are not the most effective and it shows which methods are the most useful. And few people know that using a spaced repetition system can improve your learning. Apart from learning, there are probably other areas ways you can improve your chances of success such as becoming more conscientious33 or by building a network.

So don’t see this evidence on the link between GMA and success as a reason to give up – instead use it to understand that a major cause of success is a mixture of knowledge and the ability to think, and find ways to improve your ability at those things. And then look for other major determinants of success and improve on those. In the future, we’ll write more on other determinants of success and how to improve your chances.


You may also enjoy


Highlights from the comments:

We previously had a comment section on the blog. Here are some of the highlights from the old section:

Eric Gastfriend:

It would be great to update this article to incorporate or address this criticism of IQ testing that was published in 2015.

Richardson, Ken, and Sarah H. Norgate. “Does IQ really predict job performance?.” Applied Developmental Science 19.3 (2015): 153-169.

Pablo Stafforini:

The thesis that, beyond a certain threshold (~120), IQ confers no additional benefits has been the subject of many popular books, most notably Malcolm Gladwell’s Outliers. However, the thesis seems to be contradicted by the available evidence:

  • Randomly selected eminent scientists have IQs far above the population mean (which we can approximate as the mean of the population of PhD science students: 130).
  • SAT-M scores within the top 1% of the population predict future scientific success.
  • The IQ scores of participants in the Terman study of gifted individuals (who had minimum IQ of 134) were positively correlated with lifetime earnings.

We also linked to this article from Slate Star Codex arguing that individuals shouldn’t worry too much about their scores.

Notes and references

  1. : Gottfredson, Linda S. The general intelligence factor. Scientific American, Incorporated, 1998.

  2. : There is a theory of cognitive ability called extended Gf-Gc theory which uses factor analysis to conclude that there are approximately 8 major cognitive abilities that can’t be summarised into one general factor. See Horn, John, and Hiromi Masunaga. “A Merging Theory of Expertise and Intelligence.” (2006). in Charness, Neil, Paul J Feltovich, and Robert R Hoffman. The Cambridge handbook of expertise and expert performance. Cambridge University Press, 2006.

  3. : Gottfredson, Linda S. The general intelligence factor. Scientific American, Incorporated, 1998.

  4. : When performance is measured objectively using carefully constructed work sample tests (samples of actual job tasks), the correlation (validity) with intelligence measures is about .84 – 84% as large as the maximum possible value of 1.00, which represents perfect prediction. When performance is measured using ratings of job performance by supervisors, the correlation with intelligence measures is .66 for medium complexity jobs (over 60% of all jobs). For more complex jobs, this value is larger (e.g. .74 for professional and managerial jobs), and for simpler jobs this value is not as high (e.g. .56 for semi-skilled jobs). Another performance measure that is important is the amount learned in job training programs (Hunter et al., 2006). Regardless of job level, intelligence measures predict amount learned in training with validity of about .74 (Schmidt, Shaffer, and Oh, 2008). From: Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  5. : Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  6. : Occupational level is a reasonable measure of success: “People’s rankings or ratings of the occupational level or prestige of different occupations are very reliable; correlations between mean ratings across studies are in the .95 to .98 range, regardless of the social class, occupation, age, or country of the raters (Dawis, 1994; Jensen, 1980, pp. 339–347)” From: Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  7. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  8. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  9. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  10. : Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  11. : Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  12. : Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  13. : I haven’t seen a definition of job complexity. In these studies, jobs are divided into several broad categories of complexity based on a judgement of their complexity.

  14. : Hunter, John E. “Cognitive ability, cognitive aptitudes, job knowledge, and job performance.” Journal of vocational behavior 29.3 (1986): 340-362.

  15. : Hunter, John E, Frank L Schmidt, and Michael K Judiesch. “Individual differences in output variability as a function of job complexity.” Journal of Applied Psychology 75.1 (1990): 28.

  16. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  17. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  18. : O’Reilly III, Charles A, and Jennifer A Chatman. “Working smarter and harder: A longitudinal study of managerial success.” Administrative Science Quarterly (1994): 603-627.

  19. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  20. : Schmidt, Frank L, and John Hunter. “General mental ability in the world of work: occupational attainment and job performance.” Journal of personality and social psychology 86.1 (2004): 162.

  21. : Page 11 of Schmidt, Frank L, and John E Hunter. “Select on intelligence.” Handbook of principles of organizational behavior(2000): 3-14.

  22. : Horn, John, and Hiromi Masunaga. “A Merging Theory of Expertise and Intelligence.” (2006). in Charness, Neil, Paul J Feltovich, and Robert R Hoffman. The Cambridge handbook of expertise and expert performance. Cambridge University Press, 2006.

  23. : Horn, John, and Hiromi Masunaga. “A Merging Theory of Expertise and Intelligence.” (2006). in Charness, Neil, Paul J Feltovich, and Robert R Hoffman. The Cambridge handbook of expertise and expert performance. Cambridge University Press, 2006.

  24. : “At one extreme, some fields are characterized by relatively early peaks, usually around the early 30s or even late 20s in chronological units, with somewhat steep descents thereafter, so that the output rate becomes less than one quarter the maximum. This agewise pattern apparently holds for such endeavors as lyric poetry, pure mathematics, and theoretical physics, forexample (Adams, 1946; Dennis, 1966; Lehman, 1953a; Moulin, 1955; Roe, 1972b; Simonton, 1975a; Van Heeringen & Dijkwel, 1987). At
    the contrary extreme, the typical trends in other endeavors may display a leisurely rise to a comparatively late peak, in the late 40s or even 50s chronologically, with a minimal if not largely absent drop-off afterward. This more elongated curve holds for such domains as novel writing, history, philosophy, medicine, and general scholarship, for instance (Adams, 1946; Richard A. Davis, 1987; Dennis, 1966; Lehman, 1953a; Simonton, 1975a). Of course, many disciplines exhibit age curves somewhat between these two outer limits, with a maximum output rate around chronological age 40 and a notable yet moderate decline thereafter (see, e.g., Fulton & Trow, 1974; Hermann, 1988; McDowell, 1982; Zhao & Jiang, 1986).” From Simonton, Dean K. “Age and outstanding achievement: What do we know after a century of research?.” Psychological Bulletin 104.2 (1988): 251.

  25. : Schmidt, Frank L. “The role of general cognitive ability and job performance: Why there cannot be a debate.” Human performance 15.1-2 (2002): 187-210.

  26. : Horn, John, and Hiromi Masunaga. “A Merging Theory of Expertise and Intelligence.” (2006). in Charness, Neil, Paul J Feltovich, and Robert R Hoffman. The Cambridge handbook of expertise and expert performance. Cambridge University Press, 2006.

  27. : Hunt, Earl. “Expertise, talent, and social encouragement.” The Cambridge handbook of expertise and expert performance (2006): 31-38.

  28. : Coward, W Mark, and Paul R Sackett. “Linearity of ability-performance relationships: A reconfirmation.” Journal of Applied Psychology 75.3 (1990): 297.

  29. : “One can also question the range of jobs across which these findings can be generalized. As noted earlier, neither the high (professional and managerial) nor the low (simple, repetitive jobs) ends of the occupational spectrum are represented in the present study” From: Coward, W Mark, and Paul R Sackett. “Linearity of ability-performance relationships: A reconfirmation.” Journal of Applied Psychology 75.3 (1990): 297.

  30. : Gladwell, Malcolm. Outliers: The story of success. Little, Brown, 2008.

  31. : Hunter, John E, Frank L Schmidt, and Michael K Judiesch. “Individual differences in output variability as a function of job complexity.” Journal of Applied Psychology 75.1 (1990): 28.

  32. : Dunlosky, John et al. “Improving Students’ Learning With Effective Learning Techniques Promising Directions From Cognitive and Educational Psychology.” Psychological Science in the Public Interest 14.1 (2013): 4-58.

  33. : Barrick, Murray R, and Michael K Mount. “Select on conscientiousness and emotional stability.” Handbook of principles of organizational behavior 15 (2000): 28.