Thursday, September 24, 2009

THE ETHICS OF OUR TIME VII

STATISTICS, COINCIDENCE, PREDICTIVE MODELLING

Statistics may seem straightforward. It’s just math, right? Well, yes, but…

For example, the number of births today is a statistic. Using this number, planners can take appropriate action in regard to future schooling resources. But statistics can also be deceptive; to say a certain number of people believe in UFOs may be very misleading. First of all, what, exactly, was the question these people answered? “Do you believe there is such a thing as an Unidentified Flying Object?” is quite different from “Do you believe alien spaceships are visiting Earth?” Depending on exactly how the question was phrased, this statistic could lead one to believe that a lot more people believe in aliens visiting our planet than actually do so. When you hear that a certain number of people do something or believe something (a much different case than just counting births), you should lend it no credence until you get to see the survey questions and responses for yourself.

On the other hand, statistics can clarify just how “magical” a coincidence is or is not. The Birthday Paradox (a mathematically-proven principle) states that in any random gathering of just 23 people, there is a fifty-fifty chance that at least two people have the same birthday. So if you’re sitting next to someone and find out you share a birthday, this is not a “magical” coincidence. It’s just an example of probabilities. When you speak to someone at an airport, and it turns out you have an acquaintance in common, this is not as unexpected as one might think. You both live or know people in both the city you’re in and the city where you’re headed. You are of a close enough social match to be speaking to each other. How many acquaintances do you have? How many do they have? It’s not as outlandish as it might first appear that you should have an acquaintance in common. When something seems like an amazing coincidence, it’s usually not as amazing as it might seem.

CONSPIRACY THEORIES

First, please accept the my proposition that all work is only 75% successful. That is, that there’s a 25% incompetence rate. It’s only a theory, and you can modify it upwards or downwards if you wish, but the fact is that incompetency and indifference exist, in every undertaking.

The idea that there is a network of superspies, supercomputers and superhuman agencies directing things is, to say the least, a little paranoid. The Occam’s Razor principle is a good one for this issue: “entities should not be multiplied unnecessarily.” In other words, the simplest explanation is usually the most likely. How likely is it, for example, that a hobo in a tinfoil hat is being tracked and even having his thoughts influenced by the CIA? Just keeping track of him would require:

To cover 21, 8-hour shifts (in one week) would require at least four full-time people (since each will only work 5 shifts per week), plus another person to cover sick-days, personal-leave, and vacation time. If each is paid $35,000 a year, the budget thus far is $175,000 per year. Now add in the cost of health insurance, office space, communications equipment, computers and office machines, and supplies like paperclips and toilet paper, and vans full of spy equipment. Add a manager to be in charge of the people and an administrative assistant to be in charge of everything else. Let’s say the admin person is only paid $30,000, but the manager makes $50,000. In wages alone we’ve already exceeded a quarter of a million dollars per year. Depending on the location of the office and the level of technology involved, the cost to track one hobo would be at least half a million dollars per year, and could be as much as 10 million. What does this hobo know that could possibly be worth this expense? And if you factor in the idea that not everyone does their job well at all times, how much of the time would they really be paying attention, anyway?

I tend to discount conspiracy theories simply because it seems so unlikely that very rich and powerful men are secretly working in concert, never betraying one another, never revealing their secrets to anyone, ever, without any disagreement, toward some vague goal of making themselves even richer and more powerful. Given the general disagreeableness and incompetence of people in general and rich, powerful people in particular, it just seems unlikely. I think it’s more likely your check got lost in the mail than that someone stole it, yet left all your other mail untouched (as though they had x-ray vision). And (with apologies to all hobos everywhere) I don't think the hobo knows anything the rest of us want to know badly enough to spend a dollar on it.

No comments: