It's the author's own fault. He/she starts off with "(think Facebook, Twitter, Google, etc.)", then proceeds directly to show the survey reflects a variety of industries. There are vastly different considerations for real world business versus Internet based ones, and that affects the profiles of founders for them. (e.g. PlaySpan.com founder was age 10!)

They count, and it is an interesting report when viewed through a macroeconomic lens, but a guy who starts a plumbing company or a screen printing shop is in a different situation than a couple Stanford PhD's working on a highly scalable business. The former is extremely valuable, but just not what people who frequent this site think of when they think of starting a company.

The study results would have only been surprising if they were true for web/tech entrepreneurs, otherwise it's all pretty much expected.