The In-Law Finding Everyone Quotes

There is a finding about in-laws that has been repeated so often it has stopped being treated as a finding. It goes like this. When a husband is close to his wife’s family, the marriage is 20% less likely to end. When a wife is close to her husband’s family, it is 20% more likely to end.

It is a good line. It sounds like it explains something you have felt. We went looking for the paper.

The paper exists. Half of that sentence is in it, word for word. The other half is not in it at all, and a later analysis of the very same couples found that neither half held up.

What the study did

In the spring of 1986, researchers went to the marriage licence office in Wayne County, Michigan, and recruited 373 couples who were applying. Everyone was marrying for the first time. Every couple was same-race, 199 Black American and 174 White American, and every wife was under 35.

Those couples were then followed for sixteen years, with interviews at years one, three, seven and sixteen. By 2002 the researchers knew the marital status of 99% of the people they started with. Just under half were still married. Just over 46% had divorced or separated.

In the first year, each spouse was asked how close they felt to their in-laws. One question, four points, from not at all close to very close. That single answer, given in 1986, is the whole of what the famous claim rests on.

What it found

For husbands, the claim is accurate and the number is real. The paper says it plainly: the odds of divorce fell by about 20% the closer a husband felt to his in-laws early on. That held for Black and White husbands alike.

For wives, three things went wrong on the way to the headline.

The main effect ran the other way. Before race enters the model, a wife feeling close to her in-laws was associated with lower odds of divorce, by about 18%. Protective, in the same direction as husbands, and statistically significant. The popular version has the sign backwards.

The harm was specific to one group. The reversal only appears once the analysis allows the effect to differ by race, and then it applies to White American wives. For Black American wives, who were the larger half of this sample, closeness to in-laws went on predicting a lower chance of divorce. The paper states this restriction directly.

There is no 20% anywhere on the wife side. That figure is the husbands’ number, mirrored. The significant term in the wives’ model is an interaction, and the White-wife estimate on its own does not carry a significance marker in the table it comes from.

One more thing that never travels with the headline. Conflict with in-laws predicted nothing, for either spouse. Not the closeness people worry about, but the arguing, and it did not move the odds of divorce at all.

Then the same couples were analysed again

In 2021, a different team went back to this cohort with the analysis the original paper had not been able to run. Rather than modelling husbands and wives separately, they modelled them as couples, both reports at once, 355 pairs, same sixteen years.

The headline effect disappeared.

Neither spouse’s own sense of closeness to the in-laws predicted divorce. Not his closeness to her family, not her closeness to his, not time spent with them, not whether the family was called on for support. Every one of those estimates sat against the line of no difference.

What did predict divorce was whether the two of them agreed. Couples who saw his closeness to her family the same way were about 46% less likely to divorce over those sixteen years.

The concrete version is stark. There were 36 couples in which the husband thought he was close to his wife’s family and his wife thought he was not. Twenty-six of those 36 marriages ended. That is 72%, against 48% in the 25 couples where the mismatch pointed the other way.

What these studies cannot tell you

Closeness was one question, asked once, forty years ago. The authors of the first paper say so themselves, calling the single-item measures not ideal. A four-point answer given in 1986 is being asked to stand in for a whole relationship with a whole family.

Nothing here measures change. In-law ties were recorded in year one and never again. Whether a relationship with a mother-in-law that improved over a decade matters is not a question this design can answer.

The sample is one county, one year, one kind of couple. Detroit-area, first marriages only, same-race couples only, wives under 35, and only Black and White Americans. No remarried couples, no interracial couples, no same-sex couples, no immigrant families where living with a spouse’s parents is ordinary. Divorce norms in 1986 Michigan are not divorce norms now.

The wife finding is a fragile kind of finding. It is an interaction between race and a single-item predictor, in a model with several such interactions. Those are exactly the results that tend not to replicate, and this one did not replicate when the same data were analysed as couples.

The re-analysis is not a knockout either. Its discordance categories are small — 36 couples in one cell, 25 in the other — and it is the same cohort, so it is not independent evidence. It is a better analysis of the same limited information, not a new study.

We could not find a meta-analysis on in-laws and couple outcomes. Searching the citation indexes turned up no systematic review of this question at all. Alongside money or communication, the evidence base here is thin.

What this means for your fight about his mother

The useful thing in all of this is not the ranking of who should be close to whom. It is the discovery that the question the data actually answers is a different question from the one people ask.

Couples do not usually argue about a closeness score. They argue about descriptions. He says he sees his parents a normal amount. She says he drops everything the moment they call. She says she has always made an effort with his mother. He says she goes quiet the second his mother is in the room.

Only one of those pairs of sentences can be the shared account of what happens, and in this cohort the couples who did not have a shared account were the ones who ended.

That does not mean the disagreement caused the divorce. A husband who is drifting away may well be the sort of husband who overestimates how close he is to everyone. But it does redirect the effort. Arguing about whether your mother is too involved is a debate with no evidence in it. Working out why the two of you describe the same Sunday lunch so differently is at least a question with an answer.

And the thing nobody quotes deserves saying again. Conflict with in-laws predicted nothing here. A tense relationship with your husband’s family, on this evidence, is not the threat. Being two people with two different stories about it might be.

Where a session comes into this

The session that fits this is The Same Fight, From the Outside. It takes about twelve minutes and you do it on your own.

You tell Elena the last in-law argument as you experienced it, without being fair about it. Then she asks you to tell the same afternoon again as someone standing in the doorway who wants you both to come out of it all right. What usually surfaces is not who was right about his mother. It is the two different Sundays the two of you were in.

Elena is an AI coach rather than a therapist, and she has no view on your mother-in-law. What she can do is hold you to one account of the day for long enough that you notice where yours and your partner’s come apart.