Couples where either partner grew up as an only child face a notably higher risk of divorce than couples where neither partner did, a finding that directly contradicts decades of popular birth-order compatibility advice sold in books and magazine quizzes. The analysis, from demographers at Columbia University and the Max Planck Institute for Demographic Research, examined 235,208 first marriages using Swedish population register data and found couples where both partners were only children had a divorce rate roughly 15% higher than couples where neither partner was an only child.
The paper, “Birth Order Pairings and Romantic Success,” was written by Seymour Spilerman of Columbia and Kieron Barclay, now at Stockholm University and formerly of Max Planck. Because Sweden maintains detailed multigenerational population registers, the researchers could reconstruct sibling structure and marriage outcomes for entire birth cohorts, people born between 1950 and 1990 who married between 1990 and 2012, without relying on the self-reported, memory-dependent surveys that most birth-order research has historically had to use. Restricting the sample to native-born Swedes with Swedish-born parents, no half- or step-siblings, and no twins in the family let them isolate birth-order effects with unusual precision. Across the full sample, 31,272 of the tracked marriages ended in divorce.
The finding runs directly counter to a strand of pop psychology that has circulated in various forms for decades: the idea, popularized by authors including Kevin Leman and Linda Blair, that only children make especially easy or especially difficult romantic partners depending on who they’re paired with, and that savvy daters should seek out particular birth-order matches for compatibility. Spilerman and Barclay tested these popular claims directly against their register data and reported finding little basis in fact for the specific pairing recommendations in that literature, despite how widely the advice has circulated and how confidently it’s often repeated.
What the data does support is a more general pattern: growing up without siblings appears to correlate with a somewhat higher risk of marital instability later in life, for reasons the register data can describe but not fully explain. One plausible mechanism researchers in this area point to is that siblings function as a kind of early training ground for the negotiation, compromise, and conflict-resolution skills a marriage also demands: sharing a room, competing for parental attention, working out fights without adult referees. An only child reaches adulthood, and a marriage, without that particular category of practice, which may show up later as a somewhat higher struggle rate when a marriage’s inevitable conflicts arise.
The study’s second major finding involves birth order rather than only-child status specifically: men married to first-born women showed a modest protective effect against divorce compared with men married to last-born women, relative to an only-child baseline: roughly 89% of marriages to first-born wives were intact at comparison points where about 93% of marriages to last-born wives had ended, in the paper’s survival analysis. Notably, the same pattern didn’t hold in reverse. Women who married first-born men saw no comparable protective effect, meaning whatever is driving the first-born effect isn’t simply “first-borns make more stable spouses” in general, it’s specific to first-born women as partners for men, not a universal rule that a birth-order compatibility book could package into simple pairing advice either direction.
It’s worth being direct about the paper’s limits. It’s a working paper, posted through the Max Planck Institute’s working paper series and not, as far as public records show, since published in a peer-reviewed journal, and it draws on Swedish register data specifically, a country with different family patterns, divorce laws, and average family sizes than the United States. Whether the same 15% gap would show up in an American, Japanese, or Brazilian population is genuinely unknown, and the researchers don’t claim otherwise. What the data can support is narrower but still meaningful: with a sample this large and this clean, the specific birth-order compatibility formulas sold in popular relationship advice don’t hold up, even as being an only child does appear to correlate with somewhat higher divorce risk on its own.
That’s a useful thing to know regardless of where you fall in your own family’s birth order. The popular birth-order matchmaking industry has spent decades selling the idea that who your partner’s older or younger siblings were says something predictive about your marriage. The actual data, when someone finally had 235,000 real marriages and a few decades to check, mostly says: not really, except for the one group the compatibility books rarely single out for special caution, which is the group that didn’t have any siblings to practice on at all.
