Birth Timing Does Not Predict Adult Personality, Stress or Couple Outcomes: Evidence from Four US Cohorts

Background. Season of birth, perinatal photoperiod and relative age in school are associated with psychiatric diagnoses and ADHD at the population level. Whether these birth-timing variables also shape adult personality, stress or couple relationships, and so deserve a place in relationship-risk models, has rarely been tested. Methods. Tests in four large US studies that follow the same people over many years, each planned and written down before the data for that outcome were analysed: the Health and Retirement Study (HRS; N ≈ 20,600 with birth month), Add Health (Waves I, IV, V; N ≈ 4,900), NLSY79 (N ≈ 6,800 with personality data) and the NLSY79 Children and Young Adults (CYA; N ≈ 11,500 children, compared between siblings). Together, the analytic samples include about 44,000 people. For season of birth, we tested whether outcomes rise and fall in a regular pattern over the year of birth. For relative age, we compared people born in January, the oldest in their school class, with those born in August, the youngest. These months keep the same place in class whatever date a US state used for school entry. Bayes factors (BF01) showed how strongly the data support the absence of an effect. Results. Season of birth was not linked to any of the Big Five personality traits, moodiness, anger, hostility, depressive symptoms, life satisfaction, childhood hyperactivity, childhood temper outbursts, stress, relationship happiness, or separation and divorce. For several outcomes the data actively supported the absence of any effect: they were up to about 21 times more likely if season of birth has no effect at all than if it has even a small effect (largest BF01 = 20.8, for agreeableness). Temperature in the first three months of life or during pregnancy, linked from historical climate records, also predicted no adult trait. An apparent seasonal pattern in infant temperament disappeared when age was controlled to the month. Relative age was real at school age (late-born children were held back far more often) but when the results of all four studies were combined, it was not linked to adult neuroticism (+0.02 SD, one-sided p = .12), and the studies disagreed with each other (p = .02). Relative age also did not predict an ADHD diagnosis or medication. Birth timing and birthplace did not predict stress, so an indirect path to couple outcomes through stress is not supported, and a partner's season of birth was unrelated to how much tension people felt in their own marriage. Conclusions. Confirming and extending earlier null results, birth timing in mid-latitude US populations showed no robust association with adult personality, stress or couple outcomes, and should not be used in relationship-risk assessment.Analysis code and aggregate output: https://doi.org/10.17605/OSF.IO/4X7RF