the will of the people

What percentage of a population has to feel a certain way for it to be “the will of the people”? Half? Maybe two thirds, just to be safe? How can you tell what the bulk of a population feels?

We read that as much as 1% of the population of Egypt (by generous estimates) is rioting for the ouster of President Mubarak. We assume, based on that, that this is “the will of the people.”

All we really know is that it is the will of 1% of the people. But our instincts tell us that the 1% of people who are rioting are an indicator of a much, much more wide-spread feeling.

And there is some sense behind this instinct. It is the product of a kind of unconscious statistical reasoning that is being applied to the situation. Unconsciously, intuitively, our instincts tell us something like this: You have to really hate someone a lot to actually riot for him to leave. So if 1% of the population feels that strongly, surely there is a much larger percentage wants him to get out, but is simply below the “riot threshold.”

In a way, you are assuming a kind of normal distribution of approval. On some level, your instincts are telling you that if only half of the people disliked Mubarak, then there is no way that a full 1% of the population would be above the “riot threshold” of hatred for the guy:

Normal Distribution

Therefore, given that 1% of the population is above the “riot threshold”, you assume that the distribution is shifted and that more than half of the population dislikes him.

Shifted Distribution

Thus, you conclude, the ouster of Mubarak is: the will of the people.

 

You may think this is a lot of “over-thinking” and analysis for a conclusion that ought to be very simple. But the real fact is that the only way that you can get from the fact that 1% of the population is rioting to the conclusion that the majority of the people have that desire is to employ this rather elaborate (if unconscious) set of assumptions.

The reason I want to bring this underlying and somewhat hidden logic to the surface, however, is to point out a problem with it.

What if the distribution isn’t normal?

Here in the United States, we live in a country where the far right is becoming increasingly radicalized. Conservative terrorist attempts have been increasing steadily over the last decade. The “narrow-casting” of the selective media—conservatives only listening to conservative opinions and liberals only listening to liberal opinions—acts like a kind of centrifugal force, pushing the people already at the tails of the distribution further to the extremes.

In that kind of world, the bulk of the middle of the bell curve gets smaller because moderates are pushed out towards the two extremes. And even if an equal number are pushed in opposite directions (half to become more liberal, let’s say, and the other half to become more conservative), you still have more people being pushed past that “riot threshold.”

If Fox News pays to have 3 million people bussed into Washington D.C. to hold up signs that say, “Get out Obama u Muslum!” then our statistical “instincts” will lead us to believe that it is an expression of the “will of the people.” That’s how it would be reported, and let’s face it: that’s how it would feel, to anyone watching that news broadcast on television.

But it would be wrong. It would be an expression of the feelings of a hand-picked, radicalized 1% of a population. It would not be an indicator of “the will of the people”.

I’m not saying that this is what is going on Egypt. I will be the first to admit that I don’t really know what is going on in Egypt.

But I do know this: any time anyone bandies about phrases like “the will of the people”, it is worth taking a much closer look.

imagex2