Hi, I'm Dave Mulder. This is my website, where I write about user experience and product design.

Originally Published: January 14, 2010

When the average lies

How often do you tell a lie?

Past scholarly work insists that, if you are anything like the average American, you are lying 1 to 2 times per day. That seems like a lot, but serious researchers have replicated the findings and mainstream media has picked it up as fact.

But the number still seems suspect. Are Americans (and maybe humans in general) pre-disposed to intentionally mislead others a few times each day?

Serota, Levine, and Boster (faculty members at Michigan State University) address this question in January 2010’s journal issue of Human Communication┬áResearch.

The researchers report on three separate studies: a national survey, a re-analysis of data from other deception studies, and another national survey with a different sample group.

Their findings confirm past work: the average (mean) of lying is somewhere between 1 and 2 times per day; however, this number is, well, meaningless.

Why? The distribution of data is not normal. Rather, it is skewed; most people do not lie on a given day, and the majority of lies are told by a handful of prolific liars. Those who do lie, lie disproportionately more often.

The chart above depicts those who indicated that they did lie at least once (which was only 40.1% of respondents). Because some people are lying 20 or more times per day (1.2% of this sample), their response skews the average of the entire group.

Skewed distributions are a great illustration of why the mean (average) is often meaningless.

It is interesting to note that this is not the first time a power distribution has been seen in this kind of study. The problem, however, is that past researchers did not treat it as an important finding. They were not looking for skewed data, and when they found it they did not realize what they had. They thought it meant something else.

To read the full study, check out: Serota et al. “The Prevalence of Lying in America: Three Studies of Self-Reported Lies” in Human Communication Research, 36 (2010), 2-25.