Callahan's Principle of Vagueness Reduction

I think I may have blogged on this topic before, but as I'm in the middle of reading a paper that does executes the manuever in question, goddamnit, I'm going to blog it again.

Let's say in your empirical work you have five measurements, each of which are somewhat vague as to what they mean. This amount of vagueness is disturbing. What should you do? Enter into detailed analysis of what each of the five really means?

If you said yes to the above, you're unlikely to make it as an academic. That would take a lot of time and thought! No, the right way to handle these five vague, ambiguous variables is simple: average them! Now you only have one vague, ambiguous variable. And 80% reduction in vagueness, achieved in about five seconds. Can't beat that for a efficient academic work.

Comments

Popular posts from this blog

Libertarians, My Libertarians!

"Machine Learning"

"Pre-Galilean" Foolishness