jump to navigation

Statistics is (are?) For Everyone October 13, 2018

Posted by Peter Varhol in Education, Technology and Culture.
Tags: , , ,
trackback

I taught statistics, to undergrad and graduate business students for a number of years.  I typically started off the courses by explaining how statistics were real life constructs, and were far more important in understanding the world than anything else they studied.  I especially loved word problems, which I thought were the epitome of real life.  They were analytical problems expressed in ambiguous words, with incomplete information, yet required a single correct answer.

Everyone got a good laugh out of that, and for the rest of the course treated me like the crazy uncle that they kept in the attic.

But the point remains valid, and important to anyone who cares about real life data, because there is a dichotomy between how statistics are taught, and how we might use them.

Yesterday I had a meeting with someone who told me of his teenage son, who studied and knew every conceivable football player, their university, and their statistics.  He pointed out to his son that he should really enjoy his AP Statistics course, because of his interest in football statistics.

But here’s the problem.  Most traditional statistics courses don’t teach like that.  Statistics courses are designed to look at uncertainty and how to manage it.  So we discuss mean, standard deviation, t-test, Chi-Square, ANOVA, and so on, confident that students will form a mental model of how uncertainly plays a central role in any data samples that we analyze.

Let me tell you something.  Students don’t care.  I know from years of experience that most students think statistics is the most useless course they are required to take.  They largely don’t want to be there, and I considered it a success if I were actually able to get them interested enough in the topic to do homework and understand what the answers meant in real life.

Today, of course, practically every decision made revolves around analytics.  But many business professionals still have trouble relating their university statistics classes to the decisions they make on a daily basis.  For these folks, statistics as a discipline, with an innate understanding of sampling, confidence, and uncertainty is divorced from the results they are presented in their analytics engines.

What’s the solution?  Make statistics relevant.  Teach Moneyball, fantasy football statistics, weather probabilities, or anything that makes it real to people who struggle with the math and its meaning.

I’m not a gambler, and I kind of shrug at the beginning of the era of legalized sports gambling.  But statistical inference and probabilities are at the heart of sports gambling.  If the field of statistics wants to remain relevant, it should start here.

Advertisements

Comments»

No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: