Wednesday, February 20, 2013

Correlation Does Not Imply Causation-Finding the Balance between Curiosity and Skepticism

Currently there is a beer commercial that often airs on televised sporting events in the US. In this commercial some of the fans in attendance convince a fellow fan that if they all point their beer bottle labels at the field goal kicker it will create a force field that will ensure the kick is successful. When the kick is successful, the fans stand up and cheer, including the newly convinced fan with a slightly bewildered look on his face. The Stevie Wonder song "Superstition" plays while a narrator says "It's only weird if it doesn't work." This commercial is a humorous illustration of the "correlation doe not imply causation" principle. As Nate Silver explains in his book The Signal and the Noise  "Just because two variables have a relationship to each other does not mean that one causes the other."

It's a simple logic error but an easy one to make. Today I heard a speaker make exactly the same kind of logic error...implying causation because between related variables...on a much more serious subject than beer labels and football kicks.

I attended a luncheon today with the members of one of my network of colleagues.  The guest speaker presented a theory about a fairly complicated series of events over the last ten years.  Of course he didn't present it as a theory but more as fact....the result of his ability to stitch together a pattern from a large amount of data.  The presentation also had a undertone...a theme really... of conspiracy.  I long ago became skeptical of conspiracy theorists, and today was no exception.

The speaker very convincingly weaved together a narrative connecting seemingly unconnected things.  Not only did he have a story to tell but he was impressive in his ability to present data to support his point of view and to imply connections among the data points.  I could tell he was having an impact on the audience and the post luncheon comments reinforced my perception...."Wow, that was amazing!"  "It's a frightening this has been happening and no one noticed" "The politicians and press don't want this revealed."

Needless to say, I left the meeting much at how easily my fellow attendees accepted his I was about the theory itself.  It wasn't that attendees found the story compelling.  What bothered me was the total lack of skepticism.  I think this to react to a well presented, data supported story, has broad applications and should be of interest to leaders. 

We all have beliefs and biases... mental models... of how the world works.  Mental models aren't good or bad.  They help us make sense of a complicated world with more data than any one person can absorb.  The problem isn't the model, it's what the model does.  It allows information in that supports our model and rejects data that doesn't fit the model.  As Silver says "The story the data tells us is often the one we'd like to hear..."

The lesson for the leader is whenever someone tells a compelling story supported by data and presented passionately...especially when it is presented passionately.....put your skeptical antenna up.  The leader also has to be willing to challenge their own mental models. It could be that the leader's reflexive skepticism is because the data doesn't fit their own mental model.  
Great leaders find the balance between curiosity and skepticism.  Refined engagement skills help in either instance...Challenge assumptions, but also inquire..."Help me understand"   "What causes you to conclude?" "I'm coming to a different conclusion", "Say more about this" Is there an alternative explanation?"


No comments:

Post a Comment