For well over a decade, I have been scrutinizing studies of the link between getting married and getting happy. With every new published study or review article, it becomes increasingly clear that the conventional wisdom – that getting married means getting happier – is just plain wrong.
The quality of the studies has been improving. Instead of just comparing people of different marital statuses at one point in time, we now have studies that follow people over many years of their adult lives as they get married or get divorced or widowed or stay single. They are asked repeatedly about their happiness (or life satisfaction). A review of 18 such studies showed quite compellingly that people who get married do not get happier.
The more problematic studies (comparing married and unmarried people at one point in time) continue to pile up, and they also fail to make the case that single people are miserable, and by marrying, they become blissful. They could not possibly show that, for methodological reasons.
Even with the usual ways such studies bias the methodology in favor of making married people look better than they really are, singles still look happy. Typically, people who stay single are happier than those who got married and then got divorced or became widowed. I have never, in all of these years, found a study in which the average happiness of the single people was on the unhappy end of the scale.
So Dan Buettner’s claim in the February/March 2013 AARP magazine just seemed wrong. Buettner, author of Thrive: Finding Happiness the Blue Zones Way, was offering advice about how to “give yourself a happiness makeover,” and one of his suggestions was to “Find Your Soul Mate.” His justification included this claim:
“Multiple studies have shown that married people are two times more likely to be happy than nonmarried people.”
I have emailed him repeatedly asking about the research supporting that claim, and never got an answer. I’ll assume he is drawing from the same research he mentions in Thrive – a study of happiness in nine European nations. Participants chose one of four options to describe their happiness:
- Very happy
- Quite happy
- Not very happy
- Not at all happy
The authors compared the happiness reports of people who were currently in a “stable relationship” (they are not talking friendship or family relationships here – just marriage or its equivalent) to those who were not. Already, that’s a problem, because the authors are putting in the married group not everyone who ever got married but only those who are currently married. (You can’t say that “if you get married, you will get happy” if you only look at the data from those people who got married and stayed married; those who got married and later divorced may have a different story to tell. When/if you are on the cusp of your own marriage, you don’t know which group you will end up in.) Plus, the married and the unmarried are different people, so any differences in happiness could be explained by any other way they differ other than in their marital status.
Those are the kinds of problems I’ve discussed many times before. Here I want to underscore another suspect practice – describing just a fraction of the data from the study in question. That’s what Buettner did when he made his claim, but he did not even get that partial reporting right.
I found the sentence from the 9-nation article that Buettner was probably relying on: “People in stable relationships report being very happy about twice as much as singles.”
That is a description of the people who checked off just one of the four possible answers. They are the people who chose “very happy” and not “quite happy” or “not very happy” or “not at all happy.”
Does the twice-as-often statement lead you to believe that the unmarried people were overwhelmingly describing themselves as unhappy? If it does, you have been misled.
Consider the example of Iceland. In that country, the difference between those currently married and all those not married in describing themselves as “very happy” is bigger than it is for any of the other eight countries: 57 percent vs. 22 percent. So how did the other 78 percent of the unmarrieds describe themselves? Most – 72 percent – said they were “quite happy.”
Across all nine countries, 84 percent of the non-married people described themselves as happy – either “very happy” or “quite happy.” Buettner’s claim that married people are two times more likely to be happy than nonmarried people” simply cannot be true. It isn’t true. Not even close.
Yet, there the claim appeared, in a magazine with one of the biggest circulations of any (more than 20 million). It was not offered just as a descriptive report (though even as such, it would have been inaccurate) but as prescriptive – find your soul mate, and you will give yourself a happiness makeover.
Here’s something significant that did not make it into the AARP story advising 20 million readers to find their soul mate, though Buettner did admit it in Thrive: The married people were already happier than those who were not married, even before they married.
Still another thing about that 9-nation study that you will not find Buettner mentioning in the AARP article: Participants also reported their satisfaction with their lives (in addition to describing their happiness). Apparently, marital status did not matter. Even using the cheater technique of comparing just the currently married to all of the unmarried, the married people were not more satisfied with their lives.
The point I am making is not specific to Dan Buettner or even to claims about getting married and getting happier. Whenever someone pretends to tell you what a study found by describing only a cherry-picked fraction of the answers to just one of the questions in the study, beware. If you knew what all of the rest of the results showed, you might come away with a very different impression – a more accurate one.
Another recent example of the bad practice of selective reporting is in Knot Yet, the report issued by the National Marriage Project that has generated lots of discussion of the relative merits of marrying earlier vs. later. (No, the National Marriage Project is not about to take seriously the possibility of living your best life by staying single.) Several key figures in the paper, including ones displaying happiness data, show just 25 percent of the responses. Other commentators have taken those selective displays at face value, never stopping to wonder whether a more complete reporting might have told a different story.
It is no surprise that the National Marriage Project wants us to believe that getting married transforms miserable single people in to blissful couples. But we deserve scientifically-sound, evidence-based conclusions, not just wishful ideology.
Unhappy couple photo available from Shutterstock