Is she or isn't she (happy)? Economists argue the existence of the "Happiness Gap" and why it shouldn't matter

They seem like such simple questions. Are women happier today than they were 35 years ago? If so, then why?

Guess what? We have no idea. While there has been research in the area including a just released paper that is being widely-reported in mainstream media, there is still no consensus from academics whether or not women are happier than they were several decades ago.

There are several reasons that we haven't conclusively answered these questions but I'll focus on two: 1) the subjective nature of happiness makes it difficult to measure, tabulate and compare; and 2) even if you put the inherent subjectiveness of the topic aside and assume the data accurately reflects women's happiness levels, statisticians cannot agree whether the results of the most recent study are statistically significant.

In addition to these two problems, and to me far more important, I think the issue really is that the wrong question has been asked. Instead of trying to figure out if women are happier than they were 30 years ago, we should be asking how can women become happier?

The Subjective Nature of Happiness

In her brilliant work, "The Happiness Myth," author Jennifer Michael Hecht explains the fundamental flaws in trying to treat happiness levels as absolutes that can be measured and analyzed--particularly when involving different people in different periods. Hecht notes that happiness is not definitive enough for the type of analysis being used on it. In other words, applying sophisticated statistical tools to subjective data will not make the findings any less subjective. Hecht further dismisses such efforts by noting that cultural biases will inevitably color any conclusions.

Economist Steven Levitt (of Freakonomics fame) added this to the debate:

There are a number of alternative explanations for [the happiness gap] findings. Below is my list, which differs somewhat from the list that Stevenson and Wolfers [the original authors of the study] present:

1. Female happiness was artificially inflated in the 1970s because of the feminist movement and the optimism it engendered among women. Yes, things have gotten better for women over the last few decades, but maybe change has happened a lot more slowly than anticipated. Thus, relative to these lofty expectations, things have been a disappointment.

2. Women’s lives have become more like men’s over the last 35 years. Men have historically been less happy than women. So it might not be surprising if the things in the workplace that always made men unhappy are now bedeviling women as well.

3. There was enormous social pressure on women in the old days to pretend they were happy even if they weren’t. Now, society allows women to express their feelings openly when they are dissatisfied with life.

4. Related to No. 3 in the preceding paragraph: these self-reported happiness measures are so hopelessly garbled by other factors that they are completely meaningless. The ever-growing army of happiness researchers will go nuts at this suggestion, but there is some pretty good evidence (like this paper by Marianne Bertrand and Sendhil Mullainathan) that declarations of happiness leave a lot to be desired as outcome measures.

Are the Results Even Statistically Significant?

Despite a degree in economics and a minor in statistics, I haven't been able to follow the debate between economists on whether the results found are statistically relevant. Any time you take a sample there is going to be a certain variance of data, the question is whether the variance in what you found and what you expected is statistically meaningful. The tools used to determine this can be rather sophisticated and the debate can move to an almost unintelligible level for non-statisticians. For example, one party in the debate said this:

...the ordered probit analysis found that the "Gender happiness gap" was not statistically significant, either in 1972 or in 2006, even at the 0.10 level

To which the authors responded with counter-reasoning about the probit analysis, the statistical relevance, the discounting of data to get to a desired result, etc. In short blah, blah, blah. The point is that smart people, very smart people, can't agree whether or not a the data backs up the claims so who are we to decide who's right? Until there is more of a consensus then (just like Barry Bonds' home run record) the study needs to carry an asterisk noting the results are in dispute.

The Study Asks the Wrong Question

To paraphrase positive psychologist Tal Ben-Shahar, the real question isn't whether women are happier than they were 30 years ago, it should be how can women become happier. Shouldn't that be the goal of all of this discussion regardless of how they felt 30, 50 or 100 years ago? NYT.com blogger Marci Alboher understands this. After discussing the recent research on this topic she digs right into how women can be happier:

I suggest that we women start dropping the ball a bit where we can get away with it. It’s what I call imperfectionism. I haven’t had a manicure or picked up an iron in years, and no one’s noticed. I delete most e-mails when I get home from a vacation (I use an autoresponder that says to e-mail me again after a certain date). Geesh, some days I don’t even manage to finish The New York Times.

Whether being imperfect raises your happiness level is a topic for another day. The point is that it is easy to get lost in the debate of the existence of a gender happiness gap and miss what is really important. How to be even happier?

We want you to be happier, so created this collection of posts. We also recommend you visit our Happiness Resource page.

Amazon link: The Happiness Myth: Why What We Think Is Right Is Wrong

Recap of story sources for this story: New York Times, Marginal Revolution article 1, Marginal Revolution article 2, Freakonomics (NYT), Shifting Careers (NYT)