is parenthood really worse than divorce? demographic clickbait in the washington post

The following is a guest post by Aaron Major.

If you’ve seen, or heard about, the Washington Post piece on having a baby being worse than death, read on. Lots of these science/social science articles come across my feed and while most of them bug me in various ways, this one has prompted me to write. Maybe it’s because I’ve got good friends who just had their first baby and, while they’re too tired and blurry-eyed to spend much time on the Facebook these days, I cringe thinking about this stuff becoming part of the many ‘having a baby’ conversations that they, and lots of folks, are having. To start with context. The article summarizes research recently published in the journal Demography as showing that having a baby reduces your happiness more than divorce, unemployment, or death of a partner. Yikes! That is one click-worthy headline. So what’s the problem with this article? A few things.

First, as the original study authors note, researchers already know (and I’m sure new parents and sit com writers would agree) that peoples’ general level of happiness goes down in the first year of having a kid. This is old news. What is supposed to be the new news is the amount that happiness goes down – 1.4 points on a 10 point scale – which is higher than the decrease that’s reported for divorce, death of a partner and unemployment. The issue isn’t that 1.4 out of 10 figure, it’s the construction of its relative magnitude – the implication being that having a baby is twice as bad as getting divorced.

The first question to ask ourselves: where do those other numbers come from? We get a link to the original Demography article to see where that 1.4 number comes from, but what about that graph? The article sources it to the Max Planck Institute (a highly reputable research center in Germany), but try finding that graph at the Max Planck Institute’s website. Go on, keep trying. Can’t find it? Because it doesn’t exist. Rather, the graph is constructed by the reporter (who, it should be noted, misstates the size of the happiness drop) from this sentence from the Max Planck Institute’s press release about that 1.4 figure: “This is notable compared to what international studies find for unemployment or the death of the partner (both with an average loss of one happiness-unit) or for divorce (minus 0.6 units) on the same scale.” Where does the Max Planck Institute get these figures? Who knows, they are not sourced. So, yes, shame on the Max Planck Institute for reporting data without attribution or citation, but double-shame on the Washington Post for not bothering to fact-check before not only repeating these numbers, but taking the time to put them into a bar chart. Notice also a detail in the Max Planck wording – “international studies.” In other words, not Germany. Is it really appropriate to compare the effects of becoming unemployed in the U.S. to having a baby in Germany? The answer is No.

So that’s the first problem: these numbers that are supposed to tell us “how bad” having a baby is are untraceable (at least to a nonspecialist) and, unless we know a lot more about them, unreliable. But there’s another problem, and that’s in the interpretation of these numbers. Think for a moment about differences in happiness as it relates to giving birth, getting divorced, the death of a partner, and becoming unemployment. One of these things does not belong here if we’re trying to make apples to apples comparisons between life events. Divorce, death and job loss are not only bad events, but they are also events that are likely preceded by periods of stress and unhappiness. Sure, some people suddenly get divorced when they are otherwise very happy, but divorce is likely the culmination of a long period of unhappiness. Thus the difference between how happy you are before you get divorced and immediately after is probably low – you were unhappy then and you’re unhappy now. Death of a partner–yes, there are tragic accidents that upset the happiest of relationships, but what about all those years-long battles with cancer that end in death? Now think about having a baby and the kind of life that precedes the choice to have a baby (and, as the authors note, for most people in rich countries, like Germany or the U.S., having a baby is a choice). That choice is likely made at a time when life is good and you’re very happy–just got that new job or promotion, just got married, just bought that new house. So it’s not surprising that the happiness difference will be greater as new parents go from feeling on top of the world to feeling tired and overwhelmed.

So, what’s the takeaway here. First, new parents or prospective parents, breathe out. The researchers behind this study drew very little significance from the size of that declining happiness figure and you should follow their lead. That shock-and-awe headline comes out of a PR person in a research institute writing an attention grabbing press release which is picked up by a journalist who, rather than performing the basic duties of a journalist (you know, read the original study, fact check) summarizes a piece of PR and then adds a bar graph.

None of this was hard to check. Yes, I have training as a social scientist, but I don’t specialize in family, fertility or demography. I’m not even particularly skilled with statistical analysis but still it’s taken me more time to write these thoughts than to arrive at them. Surely a reporter given the task of distilling scientific research can do the same.

And that brings me to the final, bigger point. One response to this is to say “well people just need to be more savvy and critical consumers of news.” Okay, fair enough, but I’m getting tired of the internet intellectual arms race; content producers work harder and harder to get us to click links and share stories and we have to get savvier and savvier to make sense of it all. How about this response: journalists need to do their jobs. This story didn’t come out of Buzzfeed, it came out of the Washington Post, an institution that at least lays claim to journalistic integrity and I suspect that it is partly because of this veneer of journalistic integrity that people are more likely to pay attention to this story than others.

Of course, what’s somewhat sad here is that all of this effort to push the findings into the media ether through attention-grabbing headlines detracts from the actually important findings that do come out of a very well-done study. Declining fertility rates, the social supports for new parents – these are important matters that deserve real conversation, or a least a conversation that is more substantive than “having a baby is worse than death.”

Aaron Major is Associate Professor of Sociology at SUNY-Albany.

Author: Dan Hirschman

I am a sociologist interested in the use of numbers in organizations, markets, and policy. For more info, see here.

5 thoughts on “is parenthood really worse than divorce? demographic clickbait in the washington post”

  1. Haven’t had a chance to read the article yet – and I don’t expect to get that time in the near future – so I have one question for people who have read the article. Can an undergrad with no statistical training read it and make sense of it?

    Most reporters don’t have any kind of statistical training and are heavily reliant on scholars to summarize their work in everyday English. There is tremendous information asymmetry. This makes it very easy for an attention hungry scholar to use the press to get attention. Science reporters can be more vulnerable to spin than political reporters, because the science reporter doesn’t expect as much deception and may lack other sources to point out the flaws in a study.

    If you expect the average reporter to be fluent enough in statistics to make their own judgments on quantitative social science research, independent of scholars hyping their findings, then you won’t get any reporters available to cover it.

    Liked by 1 person

    1. All true but I think it’s changing quickly now that information is cheaper and consumers are getting thinkier. Columbia’s program now offers a specialization in “data journalism,” and I’ve gotten the impression from Matt Ridley and other science writers that the community is pretty concerned about accuracy. If they lose credibility, they suffer even more than researchers because their income comes solely from the summaries not university patronage.


  2. Well said, Aaron. I’m all for more people learning about the results of social scientific studies, but out in the news world, almost all of our findings are just too boring for them to pay us any mind, and the interesting headlines lead to not-quite-right conclusions.

    Liked by 1 person

    1. Reading the various links again, this really does seem like a great case of science-journalism telephone. The original paper never even mentions divorce. The press release mentions it once in passing as a reference point, without citing the actual underlying research. The reporter picked up that comparison and used it in the same way as the press release (for context around an article mostly framed around fertility decisions in rich countries), and in turn someone (an editor?) decided it would be a good idea to make that comparison into a chart and then the main headline, without checking to see if the comparison was as solid as the main finding of the paper (and the main thrust of the press release and the new report).

      Liked by 1 person

    2. Tina: but look at vox.coml. The whole thing is supposed to have this scientistic aura, with the yellow highlighting, flash cards, and explainers. Or there’s IFLS on Facebook. I think there’s a lot of demand out there; we just limit supply. Nobody wants to be seen as an unserious attention seeker and we police that type of thing. Sadly, without more competition and volume among popularizers, quality stays low…


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s