Statistical Zombies

Kevin Drum has a good post on what he calls Statistical Zombies, 10 of the top mistakes people make when using statistics.  I particularly love #2:

What’s the survey error? Statistical sampling error in opinion polls is trivial compared to the error from other sources. Things such as question wording, question order, interviewer bias, and non-response rates, not to mention Bayesian reasons for suspecting that even the standard mathematical confidence interval is misleading, give most polls an accuracy of probably no more than ±15%. Example: a couple of years ago a poll asked respondents if they had voted in the last election. 72% said yes, even though the reality was that voter turnout in that election had been only 51%. Most polls and studies are careful to document the statistical sampling error, but who cares about a 3% sampling error when there might be 21 points of error from other causes?

You can’t fight fear with fear

A lot of people are upset about the TSA scanners, and I’m with them. It’s ridiculous how burdensome flying is becoming for no appreciable safety increase. The most dangerous part of flying is driving to the airport. We surely aren’t spending $8b to make that safer.

Unfortunately, a big part of the rallying cry is around “be afraid of the x-rays”. I was surprised how many of my tech friends got wrapped up in this one, even though the available data suggests otherwise. The FDA has a pretty thorough write up about the process and testing for the scanners. I do get that people, in general, aren’t interested in facts, but I was hoping that in a more educated and technical audience that wouldn’t be as true. Running around saying “be afraid of x-rays” is the same kind of scare mongering as the TSA is using to put all these ridiculous enhanced security measures in place.

Fighting fear with fear just generate hysteria and stampedes, and drowns out all the rational conversation, the one that shows just how ineffective and invasive these scanners are.

Software’s expands as does science

Bob Nystrom has an interesting post on the expanse of software as analogous to the expanding realms of science. Assembly programming as mathematics, C as physics, OOP as Chemistry, and dynamic language programming as Biology or even Sociology. I think there might be a more interesting explanation for this using Kevin Kelly’s technium model, but I’ll wait until I finish his book to reflect on that.

Do you believe in Missouri

I wasn’t yet following Fred’s blog at slacktivist when he made this original post.  It came out after the Gallup Poll on science attitudes in which only 40% of US respondents said they believed in evolution.  I think Fred sums it up nicely:

It’s hard to know what that means, exactly, to “believe in” or “not believe in” evolution. It’s like not believing in Missouri, or not believing in thermal conduction. Those two examples are a bit different from one another, but they both get at aspects of what this odd sort of disbelief entails.

“Not believing in Missouri” doesn’t affect the Show-Me State one way or another. To say that you don’t “believe in” Missouri is really to say that you deny it exists — that its existence is a fact you refuse to accept. That’s delusion No. 1. Delusion No. 2 is a corollary to that refusal — the idea that your belief or disbelief somehow makes it so. These are delusions because Missouri does, in fact, exist, and because its existence is not conditional upon your “belief” in the reality or unreality of that fact. Both of these deluded notions, I think, are a part of what many of those respondents meant when they told the pollster that they “do not believe” in evolution.

It didn’t help, of course, that Gallup framed the question with leading language about “the theory of Missouri.”

On the other hand, if someone tells you that they “don’t believe in” thermal conduction, it’s likely that they’re not so much saying they deny its existence as that they don’t understand what you mean when you say “thermal conduction.” For all their supposed disbelief, after all, they still avoid sitting on metal park benches in the winter. I suspect that something like that is the case with at least some of that 60 percent — that the more they can be led to understand this thing they don’t believe in, the less they’ll feel the need to disbelieve it.