Those who sing the praises of married life, take note. Most of the studies claiming that marriage makes people happier or healthier or less crazy or whatever — are garbage.

This is not to blame anyone. Of course we wonder about the curses and blessings of this institution that so many Americans cycle in and out of. The problem is that our questions are near-impossible to answer conclusively.

The ideal marriage experiment would involve some kind of love potion or aphrodisiacal arrow you could randomly chuck at people. Lacking that superpower, the best that researchers can do is to observe how people’s lives change before and after marriage, or to make comparisons between married couples and similar people who are single.

Both these approaches are flawed. The problem with the first idea is that it mixes up the effects of marriage and the effects of getting older. The problem with the second idea is that married and unmarried people are just too different to compare without reams and reams of psychological, biological, environmental, and genetic data.

But wait. What if you studied twins?

There’s actually a delightful body of research using twins to answer really tricky questions about the human condition. One of these papers, which will soon be published in the Journal of Family Psychology, provides some of the best evidence we have so far on a real question pondered in real bars all across America: Are we all just drinking because we’re lonely and single?

The researchers looked at nearly 2,500 pairs of male and female twins from the state of Washington, who answered questions how often and how much they usually drink.

The most surprising finding? In some cases, marriage causes people to drink more.

Studies usually show a correlation between being married and drinking less, but it’s hard to say what that really means. We suspect that heavy drinkers are less likely to get married, while straight-arrow teetotalers are probably more likely to tie the knot. That messes up the comparison.

The researchers focused on pairs of twins in which one twin happened to be married and the other twin wasn’t. Since the twins grew up together and were genetically identical, any differences in their drinking habits were probably caused by the difference in their marital statuses.

Compared to being single, getting married causes both men and women to drink less often. Marriage also causes both men and women to cut down on the number of drinks they have in a single sitting — men in particular.

The impact of marriage was not huge. The study was not precise enough to say exactly how much marriage made people cut down on their drinking habits, but we can make a rough estimate. Typically, the single twins were drinking a few times a week, usually 1-3 drinks each time. The married ones were drinking slightly less on average, perhaps one fewer drinking session a month, and one fewer drink per session.

To understand the effects of a divorce, researchers also compared pairs of twins where one was married and the other wasn’t married anymore. Getting a divorce doesn’t seem to make people drink more often, but both men and women have more drinks in each sitting.

It’s important to look separately at the questions of how often people drink and how much they drink, said lead author Diana Dinescu, a PhD candidate in clinical psychology at the University of Virginia. “They’re such different behaviors, both socially and biologically,” she said. “We know that binge drinking is very harmful, but some studies say drinking a little every day might even have health benefits.”

There are a lot of reasons you might expect marriage to make people drink less — married people tend to have more responsibilities, or they might have children, they might go out to bars less than they used to, they might nag each other not to drink so much. Most of these circumstances are not exclusive to marriage. The researchers find that cohabitation and marriage are pretty much indistinguishable in their effects on people’s drinking habits.

In fact, married male twins seem to be more likely to binge drink than their cohabiting counterparts. It’s unclear why that is. It could be that couples who are cohabiting are more likely to be in the earlier stages of their relationship, when people are on their best behavior. Or, it could be that married couples are more likely to have children, and that children drive people to drink. (Kidding — that’s uh totally implausible, right?)

No study is perfect, and there are two problems with twin studies in general.

One crucial assumption is that random factors caused one twin got married but not the other, and that these random factors were unrelated to their drinking habits. But you could imagine, for instance, that incidental differences in their upbringings — maybe a wild summer camp experience — nudged one twin toward alcohol, causing that same twin to stay single while the other got married. This would cause us to give too much credit to marriage for the differences in the twins’ drinking habits.

The other major problem with twin studies is that twins might not be representative of the general public. This study in particular looked at twins from Washington State, and the subjects were over 90 percent white. Not every twin responded to the survey either, so there might be some systematic response biases there.

These are minor concerns though. On the whole, the study offers some very convincing data confirming that being in a relationship does make you less likely to drink. We’ve suspected this for a long time, of course. “From my clinical judgment, marriage may cause shift in attitudes, a sense of a greater degree of responsibility,” Dinescu said.

Now we have some more facts to back up our anecdotal hunches.