Why Numbers Can’t Tell Us Everything

I’ve been on a tear recently about the numerical bias that is starting to take over our world. There seems to be an obsession with proving that the numbers are all we need to know, and I’m just not buying it. I’m referring to the tendency for news agencies to use statistical tidbits like the fact that, on average, only 12% of people wearing hoodies want to mug you (I just made that up). I’m referring to ESPN, which now uses numbers to tell us that, statistically, Tom Brady did just as well as Eli Manning in the Super Bowl (statistically, they may have; as a Patriots fan, I can tell you that, actually, he didn’t). Today, we as a culture believe that the value of what someone says is measured in the number of Twitter followers they have, or that test scores alone can measure a teacher’s effectiveness. What we don’t see in all of this is that if we only focus on those things we can measure, we miss learning about some pretty basic things that we can’t. Just because Ashton Kutcher has a zillion Twitter followers does not mean his influence will be lasting, significant, or even meaningful, beyond the type of influence Spuds McKenzie has today (which is: a funny trivia question 30 years after they were both overexposed).

Language education is a good place to look if you want to see how a numerical bias can lead to absurd and even counterfunctional behaviors. Those of us who have ever taught languages know that the way to learn a language is not through memorization of vocabulary lists or endless worksheets of grammar exercises. We know this because we know that language isn’t learned, or even used, in this way. The best way to learn a language is to try to practice communicating with it. However, because we have a hard time quantifying communicative competence, especially in junior high and high school, and because one of the main things you have to do in junior high and high schools is give people stratified grades, we forego communicating and focus on vocabulary tests and grammar worksheets. We do this to measure what people “know,” and because with a quiz or a worksheet we can assign a number like “8 out of 10 right.”

But, unless I missed something, the purpose of language instruction is to teach the language, not just prove that student A studied harder or is naturally better at memorization than student B. If we wanted to teach people to speak, we would teach differently, because we know list memorization doesn’t work. But we need grades, and we need to quantify learning, so that’s what we do. This is why someone who got straight As for 6 years of school-age French can show up in Paris a few years later and struggle to order a cup of coffee without sounding like the village idiot. We quantified, we stratified, and we missed the boat entirely.

Don’t get me wrong. I’m all in favor of math. Some of my best friends are statisticians, accountants, and physicists, math geeks one and all. But really—would you want one of these guys to order you a coffee next time you’re in Paris? Or would you rather have that sultry French guy/girl in the beret who can barely add three-digit numbers together order that next cup.

I rest my case.



This entry was posted in Education, Marketing, Statistics, Strategy and tagged , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.