IQ is not everything. That needs to be said clearly at the start, because this is a topic where people hear what they expect to hear rather than what is actually being argued. Emotional intelligence matters. Discipline matters. Character matters. Opportunity matters. A person with a modest IQ who works harder, stays curious and manages relationships well will outperform a high-IQ person who is lazy, arrogant and socially oblivious - and does so regularly. None of that is in dispute. What is in dispute, or rather what gets deliberately obscured by people with ideological reasons to obscure it, is the underlying empirical reality: intelligence, measured reliably and consistently across decades of research, is one of the strongest single predictors of academic achievement, occupational performance and life outcomes that psychometric science has ever identified. You can find that uncomfortable. You cannot make it untrue by calling it names.
This post is an attempt to lay out what the research actually shows, across the life span, in a way that is honest without being cruel and practical without pretending the findings do not exist.
Nature Becomes More Important as We Age
One of the most counterintuitive findings in behavioral genetics is what researchers call the Wilson Effect: the heritability of IQ increases substantially with age. In early childhood, shared family environment - the home you grew up in, the parenting you received, the neighborhood and school you attended - accounts for a meaningful portion of the variation in cognitive performance between individuals. Twins raised in the same household perform more similarly than twins raised apart. Adopted children show some resemblance to their adoptive families. The environment is doing real work in those early years.
But as people move through adolescence and into adulthood, that picture reverses. Longitudinal twin and adoption studies conducted across multiple countries converge on the same finding: by late adolescence, the heritability of general intelligence reaches approximately 0.75, meaning that roughly three quarters of the variation in IQ between adults is explained by genetic differences rather than environmental ones. The shared family environment, by contrast, drops to around 0.10 or less. What this means in plain terms is that the cognitive advantages or disadvantages a person carries into adult life are increasingly their own - rooted in their biology, expressed through the capacities they were born with and sharpened or dulled by the choices they make in the environments they seek out. Parents matter enormously in early childhood. By adulthood, the person themselves has become the dominant variable.
This is not fatalism. It is an honest account of how cognitive development works across the life span. Understanding it correctly allows for realistic educational interventions early, when the environment is most malleable, rather than expensive and largely futile attempts to close gaps through ideological pressure once the biology has expressed itself.
Researchers use the term genetic amplification to describe a related phenomenon: as people age, they increasingly seek out and construct environments that match their innate cognitive tendencies. A high-IQ child who finds books rewarding reads more, which builds vocabulary and analytical skill, which opens access to more cognitively demanding environments, which further develops the underlying capacity. A child whose cognitive profile is less oriented toward abstract reasoning may find those same environments aversive and gravitate toward hands-on, concrete, skill-based activities - which also develop real competencies, though different ones. The environment is not passive. People shape their environments, and their innate tendencies shape which environments they seek. Over a lifetime, this feedback loop amplifies genetic differences that were modest at the start. By middle age the gap between where someone started and where their biology has taken them can be substantial.
What the Research Actually Shows About Outcomes
The general intelligence factor - what researchers call g - is the most studied construct in all of psychology, and its relationship to real-world outcomes is among the most replicated findings in the behavioral sciences. The correlations are not subtle. Intelligence tests correlate with academic grades at around 0.60 to 0.70 in early education - which means a substantial majority of the variance in school performance is explained by cognitive ability. That correlation moderates somewhat at higher educational levels, where selection effects narrow the range of ability among students, but the relationship never disappears. The students who consistently perform at the top of rigorous academic environments are, overwhelmingly, the students with the highest cognitive ability. Effort matters. Study habits matter. But the raw material of intelligence is the platform everything else is built on.
In the workplace the picture is equally clear. A landmark meta-analysis by Schmidt and Hunter, covering decades of research and millions of workers across hundreds of occupations, found that general cognitive ability correlated with overall job performance at approximately 0.55 - a correlation larger than any other single predictor including personality, education, experience and structured interviews. That correlation rises substantially in complex, technical and managerial roles, where the cognitive demands are highest and the gap between strong and weak performers is most consequential. In simpler, more routine jobs the relationship is weaker, because the job itself does not demand much differentiation in cognitive processing. The job sets the ceiling; cognitive ability determines how close to it a person can get.
When statistical models pit intelligence, personality, education and work experience against each other as predictors of job performance, g consistently comes out on top. Not because other factors do not matter - they do - but because none of them predicts learning speed and long-term performance as reliably across as wide a range of contexts.
Income tells a similar story. The correlation between IQ and earnings sits at roughly 0.40 overall, which sounds modest until you appreciate that income is shaped by an enormous range of factors including family wealth, social connections, geographic location, industry choice and luck. A correlation of 0.40 between a single cognitive measure and lifetime earnings, controlling for none of those confounds, is remarkable. And the relationship strengthens with age. In early careers the variance in income is driven substantially by credential and entry-level positioning. By middle age, when careers have had time to differentiate based on actual performance, the cognitive signal becomes cleaner and the relationship between IQ and earnings tightens considerably.
The Military Got There Before the Academics Did
The United States military has been using cognitive testing at industrial scale for over a century and has accumulated more real-world validation data on the predictive power of intelligence than any research institution on earth. The Army Alpha and Army Beta tests deployed in World War I screened nearly two million recruits, providing the first large-scale empirical demonstration that cognitive assessment could reliably sort people into roles appropriate to their abilities. The Army General Classification Test in World War II refined that system and produced data on millions more. The lesson from both was the same: cognitive ability predicted training completion, performance in role and the capacity to handle increasing complexity with remarkable consistency across a population of enormous demographic diversity.
Today the Armed Forces Qualification Test - the AFQT, a subset of the broader ASVAB battery - serves as both a gatekeeper and a classification instrument. Scores below the 10th percentile disqualify enlistment entirely, because the research shows that training investment below that threshold produces negligible returns. The military's own data mandates that at least 60 percent of recruits score at or above the 50th percentile, a quality benchmark designed not for public relations purposes but because the research on training efficiency and field performance demanded it. Higher AFQT scores predict faster completion of technical training, fewer disciplinary incidents, better performance reviews at the 36-month mark and higher rates of advancement. Lower-scoring recruits who complete training do not catch up over time. The gap persists and in some cases widens as the cognitive demands of advanced roles increase.
The ASVAB is not a single test. It is a battery of ten subtests covering arithmetic reasoning, mathematics knowledge, word knowledge, paragraph comprehension, general science, electronics information, auto and shop information, mechanical comprehension and assembling objects. The AFQT score is derived from the verbal and math subtests and is used for enlistment qualification. The remaining subtests generate composite scores that route recruits into specific occupational fields - electronics, mechanical maintenance, skilled technical work, clerical and administrative roles. This is job matching at scale, grounded in decades of validity research, and it works. The system does not sort by race or gender in any intentional sense; it sorts by demonstrated cognitive profile, and that profile predicts training and job success across all demographic groups with comparable validity. The military figured out how to use cognitive assessment honestly because it could not afford to get it wrong. Lives and missions depend on putting people in roles they can actually perform.
The Bell Curve in Practice
IQ scores in the general population follow a roughly normal distribution - the bell curve - centered at 100 with a standard deviation of 15. This means that roughly two thirds of the population scores between 85 and 115, with smaller proportions at either extreme. That distribution has profound practical implications for how a society should think about education, training and workforce development, because different ranges of the distribution map onto meaningfully different cognitive capacities.
At the upper ranges - above 115 and especially above 130 - individuals process abstract information faster, hold more variables in working memory simultaneously, learn new material with less repetition and navigate complex systems with greater facility. These are the capacities that professional and technical occupations select for, and the research on why higher-IQ individuals tend to cluster in those occupations is not mysterious. The jobs demand cognitive work that higher-IQ individuals perform more reliably and more efficiently. A lawyer parsing a complex contract, an engineer modeling a structural load, a physician integrating diagnostic information from multiple systems - these tasks create steep cognitive gradients that IQ climbs more easily than any other measured variable.
At the lower ranges of the distribution, the picture is different but not bleak. Cognitive capacities that are less oriented toward abstract reasoning do not prevent meaningful, productive and well-compensated careers. Skilled trades - electrical work, plumbing, HVAC, welding, automotive mechanics, construction management - require spatial reasoning, procedural memory, problem-solving in concrete physical domains and the kind of practical intelligence that academic IQ tests were never designed to fully capture. A master electrician troubleshooting a complex commercial system is doing something cognitively demanding that many college graduates could not do. The tragedy is not that some people are more suited to trades than to universities. The tragedy is that we have built an educational culture that treats university attendance as the universal marker of success and trades as the consolation prize, which means we systematically underinvest in the training pathways that serve a substantial portion of the population better than four years of undergraduate coursework ever would.
The trades are not where people go when they cannot make it elsewhere. They are where a substantial portion of human cognitive diversity finds its most productive expression. Building an economy that understands that is more important than any diversity initiative ever launched.
IQ Is One Piece - but a Powerful One
The strongest objection to overemphasizing IQ is also the correct one: high-IQ individuals vary wildly in life outcomes, and lower-IQ individuals regularly build successful, meaningful lives through discipline, social skill, practical intelligence and sheer persistence. The research supports this. Conscientiousness - the personality trait most associated with diligence, follow-through and self-regulation - is a strong predictor of job performance in its own right, and in some studies rivals cognitive ability as a predictor of academic grades. Emotional intelligence, though harder to measure reliably, genuinely matters in roles involving leadership, teamwork and client relationships. Grit - the combination of passion and persistence that Angela Duckworth has studied extensively - predicts achievement in domains where raw ability alone is insufficient without sustained effort.
None of this contradicts the core finding. What the research shows is that in statistical models comparing g to personality traits, background variables and other predictors simultaneously, g typically retains its predictive power independently. Conscientiousness and IQ both matter, and they matter somewhat differently - IQ predicts learning speed and the upper limit of what someone can master; conscientiousness predicts whether they will apply themselves consistently enough to approach that limit. A highly conscientious person with average cognitive ability will consistently outperform a high-IQ person who coasts. But a highly conscientious person with high cognitive ability has access to a wider range of complex challenges and tends to pull furthest ahead over time. The variables are complements, not substitutes.
One of the more surprising findings in the IQ literature is the relationship between cognitive ability and health outcomes. Higher-IQ individuals are statistically less likely to smoke, more likely to exercise, more likely to follow medical advice, better at navigating the healthcare system and more likely to seek early diagnosis when symptoms appear. They are also less likely to be involved in accidents, both as drivers and in workplace settings. The correlation between IQ and all-cause mortality is well established in longitudinal data: higher childhood IQ predicts longer life, even after controlling for socioeconomic status. The mechanism appears to be partly behavioral - smarter people make systematically better decisions about risk - and partly biological, since the same genetic factors that influence cognitive development also appear to influence physiological robustness. Whatever the mechanism, the implication is sobering: the cognitive advantages that help people succeed in school and work also help them stay alive longer.
The Critics Who Cry Bias
The most persistent objection to IQ research is the claim that the tests themselves are biased - that they measure cultural familiarity or test-taking privilege rather than genuine cognitive capacity, and that group differences in average scores therefore reflect test design rather than real differences in ability. This argument has been examined exhaustively by psychometricians for sixty years, and the findings are consistent: well-constructed cognitive tests are not biased in the technically meaningful sense of the word. Bias in psychometrics has a specific definition: a test is biased if it predicts outcomes differently for different groups - if, for example, a given score predicts academic success for white students but not for Black students. Decades of predictive validity research show that cognitive tests do not do this. They predict academic performance, job performance and other outcomes with comparable accuracy across racial and gender groups. The test is doing the same job for everyone it is applied to.
The existence of average score differences between demographic groups is a separate question from test bias, and conflating the two is either an honest misunderstanding or a deliberate rhetorical move. The research on why group differences exist - and they do exist, in both directions across different comparisons - consistently points toward environmental factors including poverty, educational quality, health access and the accumulated effects of historical disadvantage, rather than toward genetic differences between groups or systematic test construction errors. Addressing those environmental factors is both morally important and likely to reduce the gaps over time. Pretending the gaps do not exist, or that pointing to them constitutes racism, does nothing to close them. It simply removes honest measurement from the toolkit of the people trying to help.
No policy of DEI mandates, CRT-inspired curriculum reform or institutional diversity programming changes the underlying distribution of g in the population. Those policies address access, representation and inclusion - all legitimate concerns - but they cannot substitute for the cognitive capacity that standardized assessment is measuring. Acknowledging this is not prejudice. It is the precondition for designing educational and workforce interventions that are actually calibrated to the people they are supposed to serve.
My Bottom Line
Understanding IQ honestly is not about dividing people into winners and losers. It is about preparing everyone for success in the environment where their particular combination of abilities, personality and drive can generate the most value. For some people that means a highly technical or academic career path where abstract reasoning is the primary currency. For others it means building a small business, earning a trade license or mastering a skilled craft where practical intelligence, spatial reasoning and physical competence are what the work actually demands. Success is not limited to one cognitive profile and it never has been. The electrician who built a company of twenty people did something cognitively demanding and economically valuable. The Marine who managed logistics in a combat environment did something cognitively demanding and consequential. The argument is not that some people matter more than others. The argument is that we serve people better when we acknowledge honestly what the research shows about how intelligence works, rather than pretending it does not matter because the truth is uncomfortable.
The smarter individuals will, on average, edge ahead in roles that demand abstract reasoning. That is what the data shows across a century of research in many countries using many methods. But society does not thrive by optimizing for that one dimension of human capacity. It thrives when everyone finds the environment where their particular strengths produce the most good - and when the institutions responsible for education, training and career development are honest enough about cognitive differences to actually help people get there.
IQ matters. It matters more with age, not less. Ignoring it does not make anyone's life better. Acknowledging it honestly, and building systems around what it actually tells us, is the beginning of actually serving the people who need the most help finding where they belong.
References
- Neisser, U., et al. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51 (2), 77-101.
- Schmidt, F. L., & Hunter, J. E. (2004). General mental ability in the world of work: Occupational attainment and job performance. Journal of Personality and Social Psychology, 86 (1), 162-173.
- Plomin, R., & Deary, I. J. (2015). Genetics and intelligence differences: Five special findings. Molecular Psychiatry, 20 (1), 98-108.
- Deary, I. J., et al. (2010). The neuroscience of human intelligence differences. Nature Reviews Neuroscience, 11 (3), 201-211.
- Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24 (1), 79-132.
- Duckworth, A. L., et al. (2007). Grit: Perseverance and passion for long-term goals. Journal of Personality and Social Psychology, 92 (6), 1087-1101.
- U.S. Department of Defense. ASVAB and AFQT technical documentation and enlistment standards. Office of the Under Secretary of Defense for Personnel and Readiness.
- Bouchard, T. J., & McGue, M. (2003). Genetic and environmental influences on human psychological differences. Journal of Neurobiology, 54 (1), 4-45.
Disclaimer: The views expressed in this post are the personal opinions of the author and are offered for educational, commentary and public discourse purposes only. They do not represent the positions of any institution, employer, organization or affiliated entity. Nothing in this post constitutes legal, financial, medical or professional advice of any kind. References to research findings, statistical correlations and demographic data are based on peer-reviewed sources and government documentation cited above and are intended to support analysis and argument. This post does not make claims about any individual's intelligence or potential, and explicitly rejects any interpretation that ranks human beings by worth based on cognitive ability. Readers are encouraged to consult primary sources and form their own conclusions.
