Longer reads
Pieces that need more room than a single question page, the statistics everyone misquotes, the arguments couples keep having, and what the research does and does not establish.
The premise that men and women need translating to each other has sold tens of millions of books. It does not survive contact with the meta-analyses. What does survive is more useful, and it has nothing to do with which sex you are.
A lab watches a couple argue for fifteen minutes and calls the outcome with over ninety percent accuracy. The number is real and correctly calculated. It is also not a prediction, and the difference is the whole story.
The framework is useful and it is largely untested, and those two facts have coexisted for three decades. The studies eventually ran. Here is what survived them and what did not.
Eighty-six researchers pooled 43 longitudinal studies and let machine learning loose on 2,413 variables. The strongest finding was not which trait won. It was an entire category of traits that lost.
Almost every couple assumes they share a definition of cheating. Very few have ever checked. The discovery that you did not is often more destabilising than the behaviour itself.
It is the most repeated number in the English-speaking world's conversation about marriage. It was never a measurement. It was a forecast, made in the 1970s, about a future that did not arrive.