Sit at any dinner party where the conversation has wandered into family, careers, or parenting, and you will sooner or later hear it. “Well, you can tell she’s a first-born — so responsible.” “He’s the classic middle child, always the peacemaker.” “Of course they’re rebellious, they’re the youngest.” It comes out as fluently as a weather report. Nobody asks for evidence; nobody questions the premise. Birth order is one of the few pieces of popular psychology that occupies the same conversational footing as star signs without anyone noticing the resemblance.

It is not just dinner-party folk theory. Walk into any large bookstore and find the parenting section: Kevin Leman’s The Birth Order Book has sold over a million copies since the 1980s and is still in print. Family therapists are trained on Adlerian frameworks where the “constellation” of birth order is considered foundational. Business magazines run profiles of CEOs that lean on “and of course, they were the firstborn.” Personality assessment vendors and corporate consultants build trainings around it. Even careful science journalists, when reaching for an example of a “well-established psychology finding,” will sometimes pull birth order out of the bin.

The empirical reality is unkind to all of that. Properly-controlled studies with sample sizes in the tens and hundreds of thousands have found that birth order has essentially no detectable effect on adult personality. The largest of these studies, with 377,000 American high school students, found birth-order differences on the Big Five personality traits with effect sizes below d = 0.02 — technically detectable because of the massive sample size, but practically indistinguishable from zero. Two independent research groups, using different populations and different measures, converged on the same answer in the same year. Follow-up replications in Indonesia and elsewhere have confirmed it.

This article is about that gap. A framework that has been popular for a century, taught in family therapy graduate programs, sold in parenting books, and invoked in leadership commentary has been comprehensively disconfirmed by modern large-N research. The disconfirming evidence has barely propagated into popular discourse. For strategists, hiring managers, and anyone responsible for thinking about people analytically — that gap is the interesting part.

Where The Theory Came From — Adler 1928, Sulloway 1996

The original birth-order theory of personality belongs to Alfred Adler, a contemporary of Freud and Jung and the founder of what is now called individual psychology. In a 1928 article titled “Characteristics of the First, Second, and Third Child” — published in a magazine called Children aimed at a popular audience — Adler laid out the framework that has persisted, with only minor variation, for nearly a century. (Adler, 1928)

In Adler’s telling, the firstborn child experiences a unique trauma: the arrival of a sibling that “dethrones” them from the position of sole attention. This dethronement shapes a personality that is conservative, oriented toward authority, anxious about loss of status, and inclined toward conscientiousness and responsibility — partly as a way of recovering parental approval. The second child grows up in the shadow of an older sibling and develops a competitive, rebellious, achievement-oriented personality, always trying to overtake. The youngest child, never dethroned, develops either ambition (in healthy cases) or pampered helplessness (in unhealthy ones). Only children, Adler argued, resemble firstborns but with their distinctive features intensified.

Adler did not present systematic data. The 1928 article and the related sections of his books are case-based, drawing on clinical practice and observation. The claims have the rhetorical structure of typology — vivid, intuitively appealing, hard to test, easy to remember. They proved enormously influential within the emerging family-therapy field. By the 1950s, “family constellation” assessment based on birth order was standard practice in Adlerian-influenced clinical training, and it remains so in some programs today.

The framework would probably have remained a soft, clinical-tradition idea if not for Frank Sulloway. In 1996, Sulloway — then a historian of science at MIT — published Born to Rebel: Birth Order, Family Dynamics, and Creative Lives. The book argued that birth order shapes not just personality in general but specifically the propensity to accept or rebel against established intellectual and political orthodoxies. Firstborns, Sulloway argued, are more conservative, more conscientious, and more likely to defend existing systems of authority. Laterborns are more open, more rebellious, more likely to be revolutionaries — in science, in politics, in social movements.

Sulloway’s data, in his telling, was an enormous historical study: thousands of historical figures, coded on their birth-order position and their stance on major scientific and political controversies (Darwinism, the Copernican revolution, the French Revolution, the Reformation). The book was reviewed in The New York Times Book Review, The New York Review of Books, and Nature; it spent weeks on best-seller lists; it gave birth order a fresh credentialed-feeling foundation that pop psychology had been waiting for.

The trouble is that very little of the modern evidence supports it.

Sulloway’s Evidence And Methodological Problems

Born to Rebel presents an enormous amount of data — coded historical figures, statistical models, and quantitative analysis spread across hundreds of pages and a large methodological appendix. To a reader unfamiliar with the underlying methods, it looks unambiguously like serious quantitative scholarship. To a reader who pays close attention to how the data was collected and coded, the picture is much less clean.

The two most damaging critiques came from political scientist John Modell and historian Frederic Townsend, both of whom attempted to replicate or extend Sulloway’s analyses and found serious problems. Sulloway’s coding of historical figures as “supporters” or “opponents” of various intellectual revolutions was performed by Sulloway and a small set of raters, and the criteria for inclusion, exclusion, and classification were extensively contested. Townsend’s reanalysis, published in Politics and the Life Sciences in 2000, found that the relationships Sulloway reported depended on specific coding decisions that could be changed in defensible ways to make the effects shrink or disappear entirely. (Townsend, 2000, DOI 10.1017/S0730938400014222)

A second category of problems was more fundamental. Many of Sulloway’s analyses used what is called a within-family or relative-rank framing — first-borns versus their own laterborn siblings — without adequate controls for family size, parental age, or the historical period. The same first-born from a family of two has a different statistical comparison group than the same first-born from a family of seven. Family size correlates with religion, social class, geography, and era — all of which are independent predictors of the outcomes being studied. Without explicit controls for these confounders, an apparent birth-order effect can be entirely a family-size effect or a cohort effect.

The most fatal problem, though, is one that Born to Rebel shares with every between-family birth-order study: it cannot tell a birth-order effect from a family-effect. If you compare “all firstborns in society” with “all laterborns in society” and find a difference, that difference can come from any of the variables that differ between small and large families — and there are many. Parents who have more children differ systematically from parents who have fewer, on income, education, religiosity, region, age at first birth, marital stability, and dozens of other factors. Comparing across families confounds birth order with all of these.

The clean way to test for a birth-order effect, and the way modern studies do it, is within families: compare siblings raised by the same parents in the same home. That comparison strips out all the between-family confounds. It is what the 2015 studies that broke the field open finally did at scale.

Damian & Roberts 2015 — n=377,000 US High Schoolers, Near-Zero Effects

In May 2015, Rodica Damian and Brent Roberts published “The associations of birth order with personality and intelligence in a representative sample of US high school students” in the Journal of Research in Personality. The dataset was Project Talent, a 1960 NSF-funded study that surveyed and tested 377,000 American high school students — making it one of the largest representative-sample personality datasets ever assembled. (Damian & Roberts, 2015, DOI 10.1016/j.jrp.2015.05.005)

The methodological design was the right one. Damian and Roberts ran both between-family analyses (comparing students across the whole sample) and within-family analyses (sibling comparisons inside the same household), with appropriate controls for sibship size, sex composition of the sibship, and age. They measured the Big Five personality dimensions — extraversion, agreeableness, conscientiousness, emotional stability, openness — plus several narrower constructs and IQ.

The results were striking in their consistency: birth order had essentially no effect on personality. The largest effect sizes across the entire study, for any personality trait, were on the order of d = 0.02 — meaning the difference between, say, the average firstborn and the average laterborn on conscientiousness was roughly two percent of one standard deviation. Statistically, with a sample of 377,000, these tiny differences were “significant” in the technical sense of crossing the p<0.05 threshold. Practically, they were noise. To put it in perspective: if you randomly selected one firstborn and one laterborn, the probability that the firstborn was higher on conscientiousness was 50.6% — almost identical to a coin flip.

The Adler/Sulloway framework predicts effect sizes orders of magnitude larger than this. If firstborns are genuinely more conservative, conscientious, and responsible — to a degree that should be visible in family life, clinical practice, and casual conversation — the expected effect size should be in the d = 0.3 to 0.5 range, not d = 0.02. The Damian and Roberts data rules out effects of even d = 0.1 with overwhelming statistical power. Whatever birth order does to personality, it does so at a magnitude too small to matter for any practical purpose.

The intelligence finding was different. Damian and Roberts did find a small but reliable firstborn advantage on cognitive ability — roughly one IQ point, consistent with several decades of earlier work that had reached similar conclusions on smaller samples. One IQ point is itself a modest effect — it would not make firstborns noticeably smarter in any setting outside of large-sample research — but it is qualitatively different from the personality null result. It is real, it replicates, and it is consistent with what is sometimes called the “confluence model” or the “resource dilution hypothesis”: with each additional child, parents have somewhat less attention, time, and verbal-stimulation bandwidth to allocate to any one child.

That distinction matters. The popular framing collapses everything into “birth order affects who you are.” The data say something much narrower: birth order has essentially zero effect on the major dimensions of personality and a small effect on objectively-measured cognitive ability that probably reflects parental investment dynamics rather than anything mystical about ordinal position.

Rohrer, Egloff & Schmukle 2015 — n=20,000+ Across Germany, US, And UK

Five months later, in November 2015, Julia Rohrer, Boris Egloff, and Stefan Schmukle published “Examining the effects of birth order on personality” in the Proceedings of the National Academy of Sciences. The paper used three different large datasets from three different countries: the German Socio-Economic Panel (about 5,500 family-sibling cases), the British Household Panel Study, and the US National Longitudinal Survey of Youth — pooling to over 20,000 adults across three nations. (Rohrer, Egloff & Schmukle, 2015, DOI 10.1073/pnas.1506451112)

The Rohrer team also used both between-family and within-family designs and controlled for the same confounds Damian and Roberts had handled. The results were essentially identical to the Damian and Roberts findings: zero meaningful effect of birth order on the Big Five personality dimensions, in any of the three countries, in either the between-family or the within-family analyses. The 95% confidence intervals were narrow enough to rule out the kind of effects the Adler/Sulloway framework requires.

The intelligence finding also replicated: a small but reliable firstborn advantage on cognitive measures (around 1.5 IQ points in their data, similar to Damian and Roberts), present in within-family designs and therefore not explainable as a family-size confound.

The two-paper, same-year convergence was unusual and decisive. Two independent research groups, using different samples, different countries, different specific Big Five inventories, and partially different methodological choices, arrived at the same answer in mid-2015: the personality version of the birth-order theory is wrong. The cognitive-ability version of the theory is real but small.

This was a moment that should have shifted the conversation. The data was clean. The sample sizes were enormous. The methods were appropriate. Two independent confirmations landed in the same year in prestigious venues. If the same situation occurred for a finding pop culture had not adopted, the field would have updated quickly. For birth order, the popular framing has barely budged in the decade since.

Botzet, Rohrer & Arslan 2021 — Indonesian Replication

Skeptics of the 2015 findings raised one reasonable concern: the data came almost entirely from Western, educated, industrialized, rich, democratic populations — the WEIRD problem that has dogged psychology for decades. Maybe the absence of birth-order effects in Germany, the UK, and the US reflects something specific about those societies — small modern families, high investment per child, low sibling rivalry — rather than a universal absence of the effect.

In 2021, Laura Botzet, Julia Rohrer, and Ruben Arslan tested this. Using data from the Indonesian Family Life Survey — a large, representative panel survey of Indonesian households — they replicated the 2015 methodology with about 7,300 individuals from a very different cultural and economic context. Indonesian families tend to be larger, less wealthy, and more traditionally structured than the Western samples; if birth-order effects on personality exist anywhere, they should be more visible in this kind of context. (Botzet, Rohrer & Arslan, 2021, DOI 10.1177/0890207020960997)

The findings replicated cleanly. Essentially zero effect of birth order on the Big Five personality dimensions in Indonesia, with the same kind of confidence intervals as in the 2015 papers. The cognitive-ability finding also replicated, though somewhat attenuated. Whatever is going on with birth order and personality, it is not a phenomenon that becomes visible in more traditional or less affluent family contexts. The null result is robust across cultures.

A separate, slightly older within-family study by Bleske-Rechek and Kelley in 2014 had reached the same conclusion using a more targeted design: pairs of biological siblings rated themselves on the Big Five, allowing direct sibling-versus-sibling comparison. Their result, with the cleanest possible test of the within-family effect, was the same null. (Bleske-Rechek & Kelley, 2014, DOI 10.1016/j.paid.2013.08.011)

What’s Honest To Say About Birth Order Now

After roughly a decade of large-N modern evidence, the honest summary looks like this.

On personality: essentially no effect. Birth order does not meaningfully predict extraversion, agreeableness, conscientiousness, emotional stability, or openness. Whatever effect exists is below the threshold of practical significance — far smaller than the effects of personality dimensions on each other, far smaller than the effects of parenting style, far smaller than basic measurement error in personality inventories. For any purpose other than running an n=300,000 statistical study, the personality version of birth-order theory should be treated as disconfirmed.

On cognitive ability: a small but real firstborn advantage. Roughly one IQ point on average — present in within-family designs, replicated across multiple large samples and cultures. The most-supported explanation is the resource-dilution and tutoring-of-younger-siblings model: firstborns get a slightly larger share of parental investment and verbal stimulation during early development, and the act of teaching younger siblings further reinforces firstborn cognitive engagement. One IQ point matters at the population level (it shifts averages perceptibly) but is invisible in any individual case — and it is not what the popular birth-order framework was ever claiming.

On educational attainment and life outcomes: small effects exist, mediated by intelligence and cohort effects, not by personality. Some studies find slightly higher educational attainment for firstborns, which the data say is most parsimoniously explained by the small IQ effect plus minor parental-investment factors, not by some birth-order-derived character trait.

On the Adler/Sulloway framework: not supported. The claim that firstborns are characterologically more conservative, conscientious, or authority-oriented, and that laterborns are characterologically more rebellious, open, or creative, was a useful clinical metaphor in the early twentieth century and a serious empirical claim in Born to Rebel. The empirical claim has not survived modern testing.

Why The Folklore Persists

A theory that has been disconfirmed in this thorough a way, by data this clean, should retreat from public discourse. It has not. Walk into any bookstore today and you can buy parenting books that present birth-order personality typology as established fact. Family therapy programs still teach it. Business magazines still invoke it. Why?

A few mechanisms reinforce each other.

Confirmation bias is unusually easy here. Birth-order typology is vague enough that almost any sibling pair can be fit to the pattern after the fact. If the firstborn turned out responsible, that confirms the theory; if not, well, the family had unusual dynamics. If the youngest is rebellious, that confirms it; if they’re a meticulous accountant, then they “broke the mold” — itself a story consistent with the theory. The framework cannot be disconfirmed by personal observation, which is the same property that makes star signs persist.

Salient anecdotes outweigh statistics. If you grew up in a family that fit the typology, that single data point — observed up close, over years, with emotional intensity — will outweigh in your intuitive memory any number of large-N studies you might read about. This is normal cognition; it is also the reason controlled studies exist.

The framework has narrative appeal. It tells a developmental story — the dethronement of the firstborn, the squeeze of the middle child, the indulgence of the youngest — that maps onto family experiences people remember. It makes complicated personalities feel comprehensible. Personality science offers similar explanatory power through the Big Five framework, but the Big Five does not tell a story; it just describes statistical correlations. People prefer the story.

The parenting and therapy industries have institutional incentives. Books, trainings, and therapeutic frameworks built on Adlerian birth-order theory cannot easily incorporate the modern evidence without undermining their own product. The incentive is to ignore or deflect, not update. The Adlerian therapy community, in particular, has continued to publish work using birth-order constructs without engaging seriously with the 2015 findings — which is what you would expect from a guild whose curriculum is built around them.

The disconfirming evidence is recent and technical. Damian and Roberts 2015 and Rohrer et al. 2015 are excellent papers but they sit behind paywalls in specialized journals, and the methodological subtlety required to appreciate why a within-family design with sibship-size controls is decisive is not part of normal scientific literacy. The findings have made it into a few thoughtful long-form articles — Scientific American ran a good summary in 2015 — but they have not penetrated the parenting-book market or the therapy-training pipeline.

The combination produces a stable equilibrium: large, clean disconfirming evidence on one side; institutional, anecdotal, and narrative pressure on the other. The disconfirming evidence has the truth on its side and not much else.

What This Means For HR, Hiring, And Leadership Selection

The reason this matters professionally — beyond the dinner-party correction — is that birth-order frameworks still surface in hiring, leadership, and team-building contexts. Executive recruiters mention candidates’ birth order. Business books profile “firstborn CEOs.” Team-building exercises occasionally use birth order as a proxy for predicted team dynamics. Some personality-assessment products include birth-order modules.

All of these uses should be retired. The data are clean enough to support that conclusion.

Birth order is not a predictor of any work-relevant personality dimension. If you are trying to predict conscientiousness, agreeableness, openness, or any other Big Five dimension that matters for hiring or team dynamics, use validated personality assessments — not birth-order proxies. The Big Five inventories have known psychometric properties, established validity coefficients for various work outcomes, and effect sizes large enough to be useful in practice. Birth order, by contrast, has effect sizes too small to be useful at all.

Birth order is not a predictor of leadership style. The “firstborn CEO” narrative is a base-rate illusion. Firstborns are overrepresented in many high-achievement populations not because of personality but because of mild sample-selection effects (older siblings often start school younger relative to cohort, get earlier credentials, and so on) and because in larger families there are simply more laterborns than firstborns per family — meaning firstborns are overrepresented in any single-child or two-child sample. Once you control for sample composition, the leadership-style differences disappear.

Birth order is not a useful diversity or team-composition variable. Some team-building frameworks suggest that “balancing” teams by birth order improves dynamics. There is no evidence for this. If you want to balance teams on dimensions that actually predict working-relationship quality, balance on cognitive style, communication preferences, or task-relevant skills. Birth order does not carry the predictive validity to justify its place in this conversation.

The clean rule for any hiring, leadership, or team-design decision: if a vendor, consultant, or trainer is selling you a framework that depends on birth-order categorization, the vendor is using a disconfirmed theory. Pick a different framework.

What This Means More Broadly About Family-Folklore Personality Claims

Birth order is one instance of a broader category of folk-psychology claims about personality: claims that have high cultural prevalence, intuitive appeal, vivid examples, but little or no support from controlled large-sample research. The category is large. Star signs and astrological personality frameworks are the cleanest example. “Left brain versus right brain” personality typing is another. The Myers-Briggs Type Indicator, despite its corporate adoption, has been criticized for decades by psychometricians for the same reasons. “Type A versus Type B” personality has degraded into a similar status. Color-based personality assessments and a wide range of pop-personality typologies sit in the same bucket.

The pattern is consistent enough to be useful as a heuristic. When a personality framework (a) has been popular for many decades, (b) generates strong intuitive recognition (“yes, that fits my brother”), (c) divides people into discrete vivid categories, and (d) cannot be tied to a body of peer-reviewed large-sample studies showing meaningful predictive validity for life outcomes, it is probably folk psychology with a marketing layer rather than science.

The contrast worth holding in mind is the Big Five framework: less colorful, less narratively satisfying, harder to fit to a single sibling — and substantially supported by decades of large-sample validation work, with established psychometric properties and replicable correlations with work, health, and relationship outcomes. The Big Five is not exciting at dinner parties. It is what the evidence actually supports.

The skill for a strategist evaluating any new personality claim is the same skill as evaluating any other behavioral-science claim. Ask three questions. First: what is the largest, best-controlled study that has tested this? Second: what was the effect size, and is it large enough to matter in any practical context? Third: has the finding replicated in independent samples with appropriate controls? If the answers are “small studies, large effects, hasn’t really replicated” — assume the claim will not survive scrutiny. If the answers are “large studies, modest effects, replicates” — the claim is probably trustworthy within the conditions it has been tested.

Birth order fails the test cleanly. The disconfirming studies are the largest, the effect sizes are vanishingly small, and the replications across cultures all converge on the same null. That is as decisive as social science gets.

The next century of pop psychology will probably keep talking about firstborns and laterborns the way the last century has. The evidence will keep saying that the personality version of the story is wrong. Both can be true at the same time, and both will be — until enough people who actually have decisions to make about hiring, parenting, or leadership stop using the framework and start asking what the data say instead.

Sources

  • Adler, A. (1928). Characteristics of the first, second, and third child. Children, 3(5), 14–52.
  • Sulloway, F. J. (1996). Born to Rebel: Birth Order, Family Dynamics, and Creative Lives. Pantheon Books.
  • Damian, R. I., & Roberts, B. W. (2015). The associations of birth order with personality and intelligence in a representative sample of US high school students. Journal of Research in Personality, 58, 96–105. DOI: 10.1016/j.jrp.2015.05.005
  • Rohrer, J. M., Egloff, B., & Schmukle, S. C. (2015). Examining the effects of birth order on personality. Proceedings of the National Academy of Sciences, 112(46), 14224–14229. DOI: 10.1073/pnas.1506451112
  • Botzet, L. J., Rohrer, J. M., & Arslan, R. C. (2021). Effects of birth order on intelligence, educational attainment, personality, and risk aversion in an Indonesian sample. European Journal of Personality, 35(4), 511–525. DOI: 10.1177/0890207020960997
  • Bleske-Rechek, A., & Kelley, J. A. (2014). Birth order and personality: A within-family test using independent self-reports from both firstborn and laterborn siblings. Personality and Individual Differences, 56, 15–18. DOI: 10.1016/j.paid.2013.08.011
  • Townsend, F. (2000). Birth order and rebelliousness: Reconstructing the research in Born to Rebel. Politics and the Life Sciences, 19(2), 135–156. DOI: 10.1017/S0730938400014222
  • Black, S. E., Grönqvist, E., & Öckert, B. (2018). Born to lead? The effect of birth order on noncognitive abilities. Review of Economics and Statistics, 100(2), 274–286. DOI: 10.1162/REST_a_00690
  • Zajonc, R. B., & Markus, G. B. (1975). Birth order and intellectual development. Psychological Review, 82(1), 74–88. DOI: 10.1037/h0076229

FAQ

But my family fits the pattern perfectly — my older sister is exactly the responsible firstborn, and my younger brother is exactly the rebellious laterborn. How can the research say there’s no effect? This is the most common response and the most natural one. Two things are happening. First, the birth-order typology is loose enough that most sibling pairs can be retrofit to it with a bit of generosity; “responsible” and “rebellious” cover wide territories, and parents who believe the typology often treat their children in ways that reinforce it. Second, individual cases will always cluster in a way that can fit any narrative — this is normal statistical variation. What the large-sample studies show is that when you average across hundreds of thousands of sibling pairs, with proper within-family controls, the systematic difference is essentially zero. Your family’s pattern is real; the pattern just is not caused by birth order. (It is caused by the specific people, their specific upbringings, and the specific dynamics in your home.) Compelling personal evidence and weak population-level effect can both be true at the same time.

What about the IQ difference — is the firstborn IQ advantage real? Yes, this one replicates. Multiple large within-family studies find that firstborns score roughly one IQ point higher on cognitive ability tests on average. The best-supported explanation is the resource-dilution model: parents have a finite amount of attention and verbal stimulation, and the firstborn gets a marginally larger share during early development. The “tutoring effect” — older siblings reinforcing their own learning by teaching younger siblings — likely contributes too. One IQ point is small but real. It is also qualitatively different from the personality claims, which do not replicate.

What about only children? Are they like firstborns, or different? This was Adler’s claim — that only children are like intensified firstborns. The modern data say no. Only children look like everyone else on the Big Five personality dimensions, with effect sizes essentially indistinguishable from zero relative to all other birth-order positions. The “spoiled only child” stereotype is folk psychology and does not survive modern measurement. (The IQ advantage findings for only children are mixed; some studies show a small positive effect, others show no effect, and the picture is complicated by the fact that “only child” status correlates with family characteristics like parental age and socioeconomic status.)

What about therapy that uses birth order — Adlerian family therapy, sibling-position work? Should clients seek out something else? This is a more nuanced question. Therapy that uses birth-order framings as one of many lenses for understanding family dynamics can still be helpful — not because the framings are scientifically valid as causal explanations, but because they sometimes function as useful narrative tools for clients to organize their own experience and identify patterns in family relationships. The danger is when birth-order theory is used as a diagnostic or predictive framework, or when it forecloses other explanations of family dynamics. A therapist who treats Adlerian sibling-position constructs as established science is making a scientific claim that the data do not support; a therapist who uses them as one possible lens among many is doing something more defensible. For consumers of therapy: a therapist’s framework is one of many tools they bring, and the empirical status of any specific tool is worth knowing about.

Is the Sulloway “Born to Rebel” finding completely dead? What about scientific revolutions? The specific claim — that birth order predicts acceptance of scientific revolutions — has not held up well in independent reanalysis. Townsend’s 2000 reanalysis showed that Sulloway’s findings depended on specific coding choices that other coders, applying defensible alternative criteria, did not replicate. The broader Sulloway argument that birth order affects openness to new ideas runs into the same problem as the personality claims more generally: when tested in modern large samples with within-family controls, the openness-personality dimension shows essentially no birth-order effect. The dramatic narrative of Born to Rebel is a great read; the empirical foundation is weaker than the book presents.

What about birth order and other outcomes — education, income, criminal behavior? Some studies find small effects of birth order on educational attainment (firstborns slightly more likely to complete higher education) and on certain non-cognitive outcomes that economists have studied (Black, Grönqvist & Öckert 2018 found small effects on leadership occupations using Swedish registry data, mediated by IQ rather than personality). These small effects are likely real but small enough that they would not justify birth-order categorization for any individual prediction. They are also fully consistent with the modern picture: a small IQ advantage for firstborns, propagating through education and career outcomes, without invoking any specific birth-order personality theory.

If birth-order personality theory is wrong, what should I use to understand family dynamics? The honest answer is that family dynamics are individual and contextual, and no clean predictive typology — whether based on birth order, MBTI, or anything else — captures them well. Useful frameworks for understanding specific family dynamics include attachment theory (well-supported empirically), the broader temperament literature (showing that children differ on stable temperamental dimensions from early infancy, and these differences interact with parenting in complex ways), and family-systems thinking that treats each family as a unique configuration rather than an instance of a typology. The Big Five personality dimensions are also useful — they are well-validated and they describe real individual differences without claiming those differences come from birth order or any other family-structure variable.

How confident are scientists in the birth-order null result — is there any chance the next big study will overturn it? As confident as social science gets. Two independent large-sample studies in 2015 (Damian & Roberts; Rohrer et al.), using different populations and different specific instruments, converged on the same near-zero effect with narrow confidence intervals. A 2021 replication in Indonesia produced the same result in a very different cultural context. Smaller within-family studies through the 2010s have consistently agreed. The samples and methods used to reach the null result are appropriate, well-powered, and rigorous. There is always some chance a future study with a novel methodology will overturn an established finding — that is how science works — but the probability that birth order has a meaningful effect on adult personality and the previous decade of large-sample work simply missed it is very low. The Adler/Sulloway framework should be treated as disconfirmed for any practical purpose.

replication-crisis birth-order personality-research parenting-myths evidence-evaluation
Share this article
LinkedIn (opens in new tab) X / Twitter (opens in new tab)
Atticus Li

Experimentation and growth leader. CXL-certified CRO practitioner, Mindworx-certified behavioral economist (1 of ~1,000 worldwide). 200+ A/B tests across energy, SaaS, fintech, e-commerce, and marketplace verticals.

About LinkedIn Newsletter