“42.7 percent of all statistics are made up on the spot.” – Steven Wright

25 May

Fun with Apple’s Share Growth Statistics

I just got done listening to last week’s Macbreak Weekly, in which Leo, Andy, Alex and Merlin discussed the recent NPD report that Apple products comprised two-thirds of all sales for personal computers costing more than $1,000. As someone who makes his living telling stories with numbers, this 20+ minute discussion over one statistic (“66%”) was highly entertaining for three reasons:

  • Tech journalists (and here I am speaking of the various online outlets that reported this stat, and not necessarily the ensuing MBW discussion) are no more numerate than their non-tech colleagues
  • The poor state of reporting on poll statistics extends to statistics of any kind
  • There is a reason why radio broadcasters don’t ‘Skype’ their audio in…

As someone who works for an organization that puts out market and opinion research numbers to the public regularly, I am accustomed to seeing our numbers mangled in print. The reason why the MBW discussion on this one number lasted for over 20 minutes is simple—because the various journalists who reported this number don’t ask enough questions. Because the primary reporting on this stat didn’t press harder to find out a few other salient points (like, what percentage of personal computers sell for over $1000, what was that percentage one year ago, and how about some actual numbers instead of share???) we are left with a number that raises more questions than it answers—a number that doesn’t tell a story.

I think Leo Laporte said the only thing we can actually surmise from such a number—Apple is growing (something, anything) and Windows PCs are not—at least within the $1000+ segment. But is it a larger share of a smaller pie?

The other thing I love about MBW, by the way, is when they attempt to extrapolate the behaviors and needs of people “outside the echo chamber” of the hard-core, Mac geek world and offer conjecture about that mythical 99% of America that seems beyond the ken of the other 1%. It almost always devolves from one piece of anecdotal evidence to another—how their grandmas and aunts use computers. It is almost like watching an episode of Star Trek: Kirk and Spock beam down to a rocky, forbidding planet (like “Montana”), wave their tricorders around, and have this discussion—“Jim, it is possible that our coming here will be perceived as a threat—their technology is far more primitive than our own. We may not be able to communicate.” “Spock, I’ll take care of this. They look and talk exactly like my Aunt Stella. I understand them implicitly.”

Cue explosives, exuent all.