Ah yes, research. You want the truth? Study the research–or so we all seem to think. But then the first thing you do, with any research, is look at how the data were compiled. And take it all with a grain of salt.

I saw a good example yesterday, in this post by Steve King of Small Business Labs. He’s looking at a really interesting new report called “The Anatomy of an Entrepreneur,” which was just published by the Kauffman Foundation. Here’s his summary:

  • Company founders tend to be middle-aged and well-educated and 95 percent of their survey respondents had completed college and 47 percent held advanced degrees.
  • Building wealth (75 percent), capitalizing on a business idea (64 percent) and owning a business (64 percent) were the top reasons given for starting a business.
  • Most had significant industry experience prior to starting their business. Also, 70 percent were married and 60 percent had at least one child when they started their business. The average respondents were middle-aged when they started their business.

Cool. I like this. Oh, and by the way, it turns out that most entrepreneurs are second children. And I’m a second child, and I had significant business experience, I was married and had four kids, I have a grad degree, and I fit the mold pretty well.

But no, wait. Steve discovers a major caveat:

It is important to note that 77 percent survey respondents are founders of high-tech companies. And although there isn’t much company data, it is likely most of the respondents started high-growth employer businesses with significant invested capital. Because of this, the survey results do not apply to the founders of typical small businesses.

Damn! Just when I was thinking I could win some arguments, and maybe even pinpoint who’s an entrepreneur and who isn’t. Instead, it turns out that, like just about every other survey, this one gives us just one particular view of one particular group of people who decided to answer the questionnaire.

Back to the drawing board. Interesting, but there you go again. Research is interesting, nice to know, but not solid enough to draw any conclusions from.

And probably better than just a wild guess. That is, if you have enough sense to take it all with healthy skepticism.

Tim BerryTim Berry

Tim Berry is the founder and chairman of Palo Alto Software and Bplans.com. Follow him on Twitter @Timberry.