The most successful and influential Americans come from a surprisingly narrow range of 'elite' educational backgrounds

The paper investigates whether graduates of a small set of highly selective U.S. colleges and universities are disproportionately represented among America’s most influential individuals, and whether the public accurately perceives this concentration. Motivated by debates about the societal influence of elites and growing educational stratification, the authors examine the educational backgrounds of leaders across politics, business, academia, science, arts, media, and the military. They pose two core questions: (1) What proportion of extraordinary achievers attended a narrowly defined group of elite institutions (including the Ivy League and Harvard University)? (2) Do members of the public accurately estimate these proportions? The study is important because prior large-scale cohort studies establish links between education and socioeconomic outcomes but cannot capture extremely rare outcomes (e.g., Nobel laureates, billionaires, U.S. presidents). This work aims to fill that gap and to inform discussions about the role of selective higher education in shaping American leadership and influence.

The authors situate their work within research on educational stratification, elite formation, and the influence of selective higher education on socioeconomic outcomes. Classic and contemporary analyses (e.g., Mills 2000; Domhoff 1967; Sandel 2020; Chetty et al. 2014, 2023; Karabel 2006; Clauset et al. 2015) discuss the concentration of power and the role of education in social mobility and elite reproduction. Previous studies show college attainment correlates with income and occupational status (Sewell & Hauser 1975; Tamborini et al. 2015), but these data rarely capture truly exceptional achievements. Some evidence indicates overrepresentation of top-school graduates among high achievers in certain fields (Brint et al. 2020; Wai 2013; Wai & Rindermann 2015), though prior samples were narrower. Econometric work (Dale & Krueger 2002) suggests limited causal effects of attending an elite school on income once selection is addressed, raising questions about mechanisms (ability, networks, signaling). The paper also connects to public perception research showing systematic misestimation of inequality and subgroup prevalence (Norton & Ariely 2011; Kraus & Tan 2015; Landy et al. 2017).

Design: Two complementary studies.

Study 1 (archival mega-analysis):

• Objective: Quantify attendance at a fixed set of highly selective institutions among 30 groups of extraordinary U.S. achievers (total N = 26,198), spanning politics, military, business, law, academia, science, arts, and media.
• Elite school definition: A list of 34 “Elite” schools created via admissions selectivity metrics based on standardized test scores (SAT/ACT for undergraduate; GMAT for business schools; LSAT for law schools), drawing from U.S. News data. For graduate degrees outside law/business, the same undergraduate elite list was used as a proxy for high GRE admissions selectivity to maintain a consistent elite set across categories. Ivy League schools and Harvard were coded separately. The list aimed to consistently capture general eliteness across domains; highly selective public or service academies (e.g., West Point) were not included to preserve consistency across all groups.
• Groups: 30 categories including Fortune 500 CEOs, billionaires, billion-dollar startup leaders, Nobel/Fields/Turing winners, National Academies (Sciences/Engineering/Medicine), American Academy of Arts & Sciences, American Philosophical Society, MacArthur Fellows, Pulitzer/National Book/National Magazine awardees, editors/writers at the New York Times and Wall Street Journal, presidents/vice presidents, senators, representatives, federal judges, four-star generals/admirals, Bilderberg and World Economic Forum participants, and Harvard University faculty (non-medical school). Brief definitions are provided for each group.
• Coding: Individuals were coded as having attended (undergraduate and/or graduate) any of the Elite 34, any Ivy League institution, and Harvard specifically. Additional categories: attended graduate school but not an Elite 34 school; attended college but not graduate school nor Elite 34; missing/unknown. Categories were mutually exclusive and summed to 100% within each group.
• Temporal considerations: Some samples extend back historically while others are contemporary; U.S. News-based selectivity data are contemporary. A supplemental analysis examined older vs. recent cohorts, finding older cohorts had higher Elite 34 attendance, implying potential upward bias for older samples. General population base rates were computed from the 2015 Current Population Survey and IPEDS: BA+ = 32.5%; Elite 34 degree = 1.9%; Ivy degree = 0.6%; Harvard degree = 0.2%.
• Ethics: Publicly available data; not human subjects research; exempt from IRB review.

Study 2 (survey of lay perceptions):

• Participants: 1,810 recruited via Amazon Mechanical Turk; N = 1,580 after exclusions (mean age 36.4 years; 54.1% male). Compensation: $1.25. Conducted on Qualtrics. Attention check required entering code “HKEQI.” The study and recruitment followed an IRB determination of exemption.
• Procedure: Participants were shown the list of Elite 34 schools and asked to estimate, for 15 selected groups (chosen for recognizability and coverage of the range of elite prevalence found in Study 1), the percentage who attended an Elite 34 school. They also completed a ranking task ordering groups by expected Elite 34 prevalence. Experimental manipulations included: (a) base-rate information (half of participants were shown population base rates for BA+, Elite 34, Ivy, Harvard), (b) order of estimating Elite 34 vs Ivy vs Harvard categories, and (c) a control condition for one-third of participants who estimated prevalence for a matched list of 34 non-elite control schools (with top-8 and top-1, University of Rochester) to test sensitivity to the perceived eliteness of the list.
• Measures: Percentage estimates per group for Elite 34 (and in conditions, Ivy and Harvard), and rankings. The Elite vs control manipulation tested whether participants treat elite lists differently from similarly recognizable non-elite lists.
• Analysis: Compare participant estimates to Study 1 actual percentages; assess under/overestimation across groups; evaluate effects of base-rate provision and control list on estimates; compare rankings to percentages for robustness.

Study 1 (archival):

• Across all 30 groups (N = 26,198), a non-weighted average of 54.2% attended one of the Elite 34 schools. Group ranges: approximately 11.2–25.9% for Four-Star Generals, Four-Star Admirals, and House members; up to 78.9–80.9% for Forbes Most Powerful Men, Harvard Faculty, and American Philosophical Society.
• Ivy League: Average 36.3% attended an Ivy League institution, ranging from 5.1% (Four-Star Generals) to 70.0% (Fields Medalists).
• Harvard: Average 16.0% attended Harvard; range 2.0% (Four-Star Generals) to 44.5% (Harvard Faculty).
• Representative mid-range examples for Elite 34 attendance: Fortune 500 CEOs 41.9%; Wall Street Journal editors/writers 50.8%; National Academy of Medicine 60.5%; National Academy of Sciences 70.5%.
• Concentration within the elite set: Top 16 elite schools account for 44.6% on average; of all Elite 34 representation, 82.3% is accounted for by the top 16, 67% by the Ivy League, and 29.5% by Harvard alone.
• Undergraduate vs. graduate attendance: Ivy League undergraduate 19.6% vs graduate 25.3%; Elite 34 undergraduate 32.2% vs graduate 39.4%; Top 16 undergraduate 23.3% vs graduate 33.7%—indicating graduate degrees from selective schools are somewhat more common.
• Overrepresentation relative to population: In the general U.S. adult population, 32.5% have BA+; 1.9% a degree from Elite 34; 0.6% from Ivy; 0.2% from Harvard. Among extraordinary achievers, Elite 34 attendance (~54%) is ~28× base rate; Ivy (~36%) is ~60×; Harvard (~16%) is ~80× (75× excluding Harvard faculty).

Study 2 (perceptions):

• Participants generally underestimated elite-school attendance among high achievers. For 10 of 15 groups, estimates were lower than actuals. Largest underestimates included: American Philosophical Society (45.7% estimated vs 80.9% actual; −35.2 pts), Harvard Faculty (65.1% vs 80.5%; −15.4), Nobel Prize Winners (45.9% vs 76.1%; −30.2), National Academy of Sciences (50.4% vs 70.5%; −20.1), MacArthur Fellows (41.5% vs 58.9%; −17.4).
• Overestimation occurred mostly for groups with low elite-school attendance: Four-Star Generals (27.4% vs 11.2%; +16.2), House members (43.8% vs 25.9%; +17.9), Federal judges (53.4% vs 41.8%; +11.6), Fortune 500 CEOs (47.1% vs 41.9%; +5.2).
• Participants shown the elite list gave higher estimates than those shown a matched control list of schools, confirming sensitivity to perceived eliteness.
• Providing population base-rate information led participants to give lower estimates across all categories, paradoxically increasing underestimation because actual elite-school attendance among high achievers exceeds intuitive expectations even without base-rate prompts.
• Ranking tasks produced orderings consistent with percentage estimates, supporting robustness of findings.

The findings demonstrate a strong link between attendance at a small set of highly selective institutions and achieving positions of exceptional influence in the United States. This directly answers the research questions by showing: (1) graduates of the Elite 34—especially Ivy League and Harvard—are vastly overrepresented among top achievers across 30 domains, and (2) the public generally underestimates the magnitude of this concentration. The results underscore the societal importance of pathways through elite higher education and raise questions about mechanisms. Potential explanations include selection on pre-college characteristics (cognitive abilities, skills, personality), educational and peer effects during college, and reputational/network advantages (including alumni networks and hiring practices favoring elite credentials). The “Harvard Effect”—Harvard’s outsized share—may reflect a combination of selection and treatment effects, including ambition and expectation-setting. Elected elites (e.g., legislators, generals/admirals) are more similar to the general population in elite-school representation, whereas anointed elites (e.g., academy members, major award winners, high-level cultural and scientific elites) show the highest concentrations, highlighting different selection and gatekeeping processes. Policy discussions about access to selective institutions and diversification of leadership pipelines are implicated, given the potential reinforcement of inequality through educational networks and signaling effects.

This study provides the first large-scale, cross-domain quantitative documentation of how concentrated elite educational backgrounds are among America’s most influential achievers and how the public perceives this pattern. The authors assembled a novel dataset spanning 26,198 individuals in 30 elite achievement categories and showed substantial overrepresentation of graduates from a small set of institutions—especially Harvard and the Ivy League. Public estimates generally fall short of the actual concentration. These social facts illuminate the structure of American elite formation and invite debate on whether leadership would benefit from broader institutional representation or greater diversity within elite institutions. Future research should: (a) incorporate additional fields and media outlets to broaden coverage; (b) examine causal mechanisms disentangling selection from treatment, human capital from signaling, and the roles of networks; (c) explore temporal dynamics using harmonized historical data; (d) analyze subgroup representation (race/ethnicity, socioeconomic background) where reliable data can be gathered; and (e) extend analyses cross-nationally to compare elite formation across education systems.

• Descriptive design: The study cannot establish causal mechanisms underlying the association between elite education and extraordinary achievement.
• Sampling and coverage: Although large and diverse (N = 26,198 across 30 groups), the selected samples may not fully represent all American high-achiever populations; choices (e.g., focusing on two major newspapers) limit scope.
• Temporal heterogeneity: Groups were compiled from differing time periods; comparisons to 2015 population base rates and reliance on contemporary selectivity measures may introduce biases. Older cohorts showed higher Elite 34 attendance, potentially inflating historical estimates.
• Measurement constraints: The consistent Elite 34 list excludes some selective public or service academies to maintain comparability across domains; this may understate domain-specific institutional influences (e.g., military academies for generals/admirals).
• Data gaps: Race and ethnicity could not be reliably coded across samples, precluding analyses of underrepresented groups.
• Mechanisms unresolved: The data cannot adjudicate between selection effects, educational/peer effects, network advantages, or signaling.
• Study 2 understanding: Participants may not fully grasp some elite groups; base-rate information led to systematic downward adjustments, potentially increasing underestimation.