Honey, I shrunk our income gap

In my freshman year of college, I frequently sat next to an extremely cute guy in my Spanish class. I was generally late, so we never arrived together, and, for reasons that now escape me, we never left at the same time either.

But we did manage to flirt a lot, and we eventually arranged to meet for a study date. It was clear that we weren't going to pay any more attention to irregular verbs on this study date than we had in class.

Earlier in the semester, our heights had come up during conversation - as they will, if you are a 6-foot-2 woman. I had given into the temptation to cheat a little. "Six feet," I lied, not daring to shave off more than that. He had replied affably, "I'm 5-10."

Now, as I prepared for the date, my aghast roommate asked me how I could consider dating a man who was two inches shorter than me.

"Because I would like to go on more than one date a year."

We were meeting in a coffee shop; he stood as I walked in. At which point it became apparent that no mere two inches separated us. The gap was more like half a foot. He too, had given into the temptation to cheat a little, adding a few inches. The sum of our little fudges was a wild distortion of reality.

And why am I sharing this story? Because of an interesting little bit of news from the Census Bureau: Couples where the wife earns more than the husband systematically overreport his income and underreport hers.

It's not an enormous gap - it looks like husbandly income tends to be overreported by about 3 percent while wifely income is underreported by 1.5 percent. But together, that could skew the data significantly, both in the number of people who are reporting that the wife earns more and the size of the gap they report.

For example, in a couple where the wife actually earns 10 percent more than the husband, the self-reported gap would cut that number almost in half.

This is, obviously, an example of a sexist double standard; in a society where males are still often defined by their ability to be "breadwinners," couples are clearly uncomfortable with new economic realities. And apparently they are frequently resolving that discomfort by lying about it - not with flagrant lies, but with little cheats that nudge them in the direction that better conforms with social norms.

The discrepancy is the latest example of how self-reported data tends to skew toward what the respondent would like the data to be, rather than what it actually is.

People overreport voting, and they underreport their weight. And some measures, like individual income, can be both overreported and underreported, with higher-income people cheating down toward the mean, and lower-income people cheating upward.

These sorts of fibs are generally harmless, of course. When the cute guy from Spanish class and I realized our joint dissembling, we had a laugh, and we went on a few more dates. But when it comes to data, the distortion can warp our understanding of the world.

Worse, it can end up reinforcing norms that are outdated - if you're part of a wife-earns-more couple, and the intentional misreporting by others makes you think your relationship is a bizarre outlier, rather than increasingly normal, maybe you'll feel strange and isolated. And then maybe you, too, will lie to researchers rather than admit you're a freak.

I could exhort people not to lie in pursuit of "normalcy" - and there I guess I just did. But an exhortation isn't going to clean up the national data problem, so here's a warning to go with it: Don't put too much faith in self-reported data that can't be verified by other means.

When you see reports about how people answered a researcher's question that might reflect on them socially in some way, be skeptical. Remember that there's always a gap between our self and the face we present to the world.

And that when multiple people are lying, they can collectively turn that gap into a chasm.

Megan McArdle is a Washington Post columnist who writes on economics, business and public policy. She is the author of "The Up Side of Down." McArdle previously wrote for Newsweek-the Daily Beast, Bloomberg View,the Atlantic and the Economist.