HR Measurement: The Difference Between Correlation and Causality
Several years ago when I worked at a big firm there was a group of people who were pushing “human capital measurement.” I got a meeting with a very important client in The Hague, and was in his office while a few HCM members called in from New York. At some point during the conversation, one of my coworkers said to the client over the conference line: “You just don’t understand your own data. We can do a sweeping survey of your people and get causation on several dimensions.”
Once I was able to pick my client off the floor…well, you know the rest. It wasn’t good. Two hints: 1) Don’t tell clients they don’t know what they’re doing, and 2) Don’t use the word “causation” to a scientist unless you really mean it.
The words correlation and causation get flung around casually in some HR circles, which is why when I read this simple summary of both terms this morning I thought it should be required reading. Here’s what Scott Selhorst at Tyner Blain writes in the opening paragraph:
One of the most common mistakes people make when looking at data is to jump to conclusions about the data. We all live in a world of cause and effect. It is only natural that when we see data that appears to show cause and effect, we assume that it does. But it often doesn’t.
Hey, it’s cold out this morning. And it’s dark. The cold makes the sun go out. Correlation? I don’t think so. It’s super important that HR people understand cause-and-effect in the business world. Use stats, but use caution when claiming that the latest training program delivered all the business results. You can be embarrassed pretty quickly.
Comments
11 Responses to “HR Measurement: The Difference Between Correlation and Causality”
Leave a Reply
-
SUBSCRIBE
-
-
-
i heard a story on NPR in the last month about this and how it’s been a real problem when reporting the news these days. papers/networks/radio have been getting correlation based data and presenting it as fact more and more and selling it to the public as such. i just did a quick search on the WHYY website for the segment, but couldn’t find it. arg.
Ah, Albert, I’d like to hear that. It makes sense…we believe things with stats even when they’re wrong!
Albert is so right (like you). One resource that I use to constantly keep myself in check is reading a column in the Wall Street Journal called “The Numbers Guy”. Each column is a detailed explanation about the proper or improper use of quantitative analysis, or data, or the description of numbers.
If you want a great exercise, look up the archives of the column. In it he has a quiz that is used by an editor to determine if writers can express numerical concepts and facts accurately in their writing. I use it with my HR team to reinforce that it’s not just about the number being correct, it’s describing to the audience what that number is correct about. What does that number mean. I am in Compensation and Benefits - we use numbers all of the time and making sure that our lay person audience really understands us is a matter of business and ethical responsibility.
My favorite way to remember the difference is to remember a teacher I had for one of my quantitative methods courses. He started the lecture by biting into a big juicy Bermuda onion, looked at the class and said, “Now I’m going to die. If you bite into a raw onion you will die.” That launched us into a discussion of causation and correlation.
Bill, I guess I’m not a WSJ guy, but I need to be. That column alone would be worth the subscription. Plus, that cool single column story that they do. Thanks for that ref.
That kind of teaching with your people is what’s essential. As you say, get the numbers right and make them make sense. Very, very good indeed.
Hi Wally — that’s a really great example. And it really gets the mind moving…I am still moving my mouth around as I ponder that one. I’ll borrow that one and give you attribution when I do.
Thanks VERY much for linking to my article. I would love to see the WSJ story too. I was inspired to write this while reading Freakonomics on the plane on Monday. Fantastic book!
Anyway, loved your article too, and I appreciate the mention.
Scott
Scott, you write some really brilliant material and I really like reading your stuff. You’re a thinker in blogdom and I’ve learned a lot about your project approach. It’s excellent.
Thanks for the article…nice work.
Great topic Frank - near and dear to my heart. Bill, thanks for the tip on The Numbers Guy.
One of my stats profs started his class like this:
“There are three kinds of lies - lies, damned lies and statistics.”
- Mark Twain
The 2008 election is coming up - all three kinds are in play.
John, that’s a god one by Mark Twain. So true. My fave quote of all time is by Mr Hoderny: Get fired with with enthusiam or you’ll be fired with enthusiasm. The best!
Unfortunately we can’t attribute that quote to my dad - that one belongs to Vince Lombardi.
The one you can attribute to my dad is, “Either be an example or I’ll make one out of you.”
John, that’s even BETTER than the Vince Lombardi line…fun. That’s the winner.