Monday, August 29, 2016

For the start of the 2016-7 academic year: On being wrong

Welcome (back) to school.  At least at Case Western Reserve University, classes start today, and for freshmen, this will be the first day of college.  Let's talk about getting things wrong.  This blog exists because of wrongness.  A year ago, I began teaching a freshman seminar by making two predictions: a random stock market drop would correct itself within a week (it did), and Donald Trump would never be president.  I currently have 80% confidence in that latter prediction, but, um, I never thought he would make it this far.  Almost none of my fellow political scientists did either.  I started this blog with a series called "Trump to Political Science: Drop Dead," because, back in February, just by maintaining his lead in the polls that long, he was already blowing our preconceptions to shreds.  We got it wrong.  Even if Trump loses to Clinton, we got the nomination wrong.

A few years ago, at a panel at the American Political Science Association honoring one of my grad school professors, Ray Wolfinger, somebody observed that Ray had the unique honor of being the only high profile political scientist ever to have the honor of never having been debunked.  I'm not Ray Wolfinger, and neither are you.  So, as a 2016-7, welcome (back) to school post, let's talk about being wrong.

1.  You are going to be wrong about stuff.  It will suck.  Get over it.  Other people have real problems.

2.  As reality becomes clear, don't be that asshole who digs deeper into denial.  Nothing is more pathetic.

3.  The down-side of being right is that you learn nothing.  If you are going to be wrong, at least get something out of it.

4.  Don't over-correct.  Sometimes a smart bet loses money.

5.  Don't under-correct.  Learn from your mistakes, or you will lose more money.

6.  Read about debunked research, etc.  That way, you can learn about other people being wrong, and learn from their mistakes.  And laugh at them.  Like, the entire field of psychology right now.  (If the election maintains stagnation, I'll probably start writing about how none of their "findings" can be replicated).

7.  Learn statistics.  Yes, I put the word, "wrong," in the title of the post, and I keep using it, but really, if I weren't writing a quick post at 8:00 before the day begins, I'd be more nuanced.  Everything is about how confident we can be, and we need a mathematical system for that.  That branch of mathematics is called "statistics."  Learn it.

8.  Go read Expert Political Judgment, by Philip Tetlock.  I reference this one a lot.  Everyone needs to read it.  It is about being wrong, who gets stuff right, why, and how to determine whether or not you were right for the wrong reasons, wrong for the right reasons, etc.

That's all for now.  Welcome back.  Get readin', kids.

No comments:

Post a Comment