Why a study about couples is not a rule about yours

A link arrives in the family group at midnight. Couples who share the chores are less likely to separate, it says, and no study is named anywhere in the piece. Or the article in your feed says the first sixty seconds of an argument tell you where the marriage will be in ten years. One of you reads it and says nothing. The other one reads it and quietly measures the week against it.

Both of you are doing the same thing, reading a sentence about strangers as a verdict on one house. It is not a verdict. It is a sentence about an average, and the three words that should slow you down are is linked to.

The three words to slow down on

Two things that arrive together in the data are not one thing causing the other. A headline uses is linked to because it is honest and because it reads like a cause. The cause is supplied by whoever reads it, and it usually arrives inside a second: they do the chores, so they must be closer, so we should do them too. The study itself claims none of that. It says two things moved together in one group of people, once.

The other half of the same trick is the study nobody names. A line borrows the authority of research and names none of it. Not which study. Not the size of the sample. Not whether the same couples were followed for years or asked once on a Tuesday evening.

So look at who was studied, and how. Relationship research is real knowledge and worth reading. Where it comes from is narrower than a headline can show. In 2010, three researchers, Joseph Henrich, Steven Heine and Ara Norenzayan, went through the psychology literature and counted where the people in those studies came from. The overwhelming majority were from Western, educated, industrialised countries, which hold a small share of the world's population, and a large share of them were university students. A finding about the average of that group is a finding about that group. It has something to say about a marriage in Kochi, but it is not a rule for one.

A single finding is a thin thing to lean on even when it is well run. In 2015, a collaboration of researchers repeated one hundred published psychology studies and reported that most did not produce the same significant result the second time, and that the effects which did survive came out about half the size of the originals. That project was about psychology as a whole rather than couples alone, and it is a reason to hold a finding as something worth trying rather than something known.

What a finding is still good for

Keep the article. Move it one shelf, from this is how it is to this is worth trying. Read as a rule, an average becomes a way of keeping score. One of you is the one who does the thing and the other is the one who does not, and the article has settled it. That is how a sentence about strangers gets a vote in your Tuesday night argument about the dishes.

Held as something to try, the same finding is useful, because it is a place to look. A finding about how couples answer each other is a hypothesis about your house, and you are the one who gets to check it.

One thing to try this week

Take one couples article that has reached either of your phones in the last month. A forwarded one counts. Tonight, after dinner, before either phone goes back on the table, open it together and answer two questions out loud, one at a time:

What is the sentence saying, together or causing?

Who was studied, and were they asked once or followed over time?

Then one question each, answered in a sentence:

Where did we do the opposite of this in our own week?

Ten minutes, once, this week. Afterwards, notice how often the first two answers are simply not in the piece, and how much smaller the claim looks once it has been said out loud.

The first time, this can feel like nitpicking, or like turning a shared article into a small audit of your partner's reading. That passes. What is left is a way of holding a claim together that neither of you has to defend.

Each Sunday: one question to ask your partner, the research behind it, and one thing to try. The free Relationship Checkup is open to everyone, any time.