Here are the main studies underlying the UK (and general) findings on degrees and intelligence:
Strong predictive link in the UK
Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007).
Intelligence and educational achievement.
Intelligence, 35(1), 13–21.
• Prospective study of >70,000 English children.
• Cognitive ability (g from the Cognitive Abilities Test) measured at age 11.
• GCSE results in 25 subjects at age 16.
• Latent correlation between g and overall educational achievement: 0.81.
• g predicted success in every subject (strongest in maths and English).
• At average g, ~58% achieved five or more A*–C GCSEs; +1 SD → ~91%; –1 SD → ~16%.
This is one of the largest and cleanest UK demonstrations that childhood general intelligence strongly predicts later educational outcomes.
Causal effect of education on IQ scores
Ritchie, S. J., & Tucker-Drob, E. M. (2018).
How much does education improve intelligence? A meta-analysis.
Psychological Science, 29(8), 1358–1369.
• Meta-analysis of 142 effect sizes from 42 datasets (>600,000 participants).
• Three quasi-experimental designs (control for prior intelligence, compulsory schooling reforms, school-entry age cutoffs).
• Additional year of education associated with 1–5 IQ points (overall estimate ≈ 3.4 points).
• Effects appear across cognitive domains and can persist into later life.
• Includes UK data (e.g., British Cohort Study analyses).
This is the key paper showing education produces modest genuine gains in cognitive test scores, not only selection.
UK higher-education expansion and graduate ability
Ichino, A., et al. (papers using Understanding Society data, circa 2022–2025).
Titles include variations of College education, intelligence, and disadvantage: policy lessons from the UK in 1960–2004 (or “Graduation nation…”).
• Tracked the large expansion of UK university places after the 1960s Robbins Report.
• Average cognitive ability of graduates declined by about 0.13 standard deviations between the 1960s and early 2000s.
• Expansion mainly pulled in lower-ability students from more advantaged backgrounds rather than high-ability students from disadvantaged ones.
• Non-graduates’ average ability also fell slightly, consistent with the new graduates sitting between the old graduate and non-graduate means.
Supporting British cohort evidence
Analyses from the 1958 National Child Development Study (NCDS) and 1970 British Cohort Study (BCS70) repeatedly show:
• Clear gradients of childhood cognitive scores by later educational attainment (no qualifications → O-levels/GCSEs → A-levels → degree).
• Cognitive ability accounts for a substantial but incomplete share of social-class differences in educational success.
• Social background effects remain after controlling for early ability.
These cohort studies provide the longitudinal backbone for UK claims about ability, education, and later outcomes.
Summary of what the studies collectively show
• Childhood general intelligence is a powerful predictor of educational attainment in Britain.
• Education itself produces small but real gains on cognitive tests.
• Expanding access reduced the average ability of the graduate population.
• Degrees therefore raise the probability of above-average cognitive ability on average, but the distributions overlap substantially and a degree is not equivalent to higher intelligence.