In my work with teams who make decisions together about tactics, strategy, and personnel, logical fallacies and cognitive biases show up, so I’m going to write a few posts about some of the most common. This is the second post in this series. You can find the first post here.
Fallacy: The False Cause, aka The Correlation Fallacy
Definition: Asserting that a relationship between two facts means that one causes the other.
Simple Example: The last two times we had a profitable quarter, we spent less on marketing. Therefore, let’s permanently cut the marketing budget.
What it looks like: Because we want to be logical, we look at facts. When we find correlations, we turn those into a cause and effect relationship. It seems logical enough, but we fail to look deeper, even though we can. A good example of this can be found between ice cream sales and murder. There is a statistically significant correlation between these, citing as ice cream sales increase, murder rates also increase. Does this mean ice cream sales causes people to murder? Of course not. There are several other variables at play.
At Work: There’s a meeting over the financials, and we need to increase profitability. I point out that when marketing costs were low, we made more money.
Someone else in the room connects the two clauses. “Marketing doesn’t work for us as well as satisfied clients and word of mouth; our former CEO always felt that way too.” (That’s the “Appeal to Authority” fallacy working its way in here also.)
The conclusion becomes: “This is clear as day; when the marketing spending is low, we have a better bottom line. Let’s cut two people from the marketing department. The other three can handle the basics. Clearly our spendy ‘razzle-dazzle’ promotions didn’t work.”
What to do?
Question causal connections that you—or others—make by asking questions such as:
“How do we know for sure that one CAUSES the other and isn’t just a coincidence?”
“Could something else have caused BOTH of these things, and we haven’t seen it yet?”
“Could we be looking too long-term, when a short-term cause might be the case?”
“Could we be looking too short-term, when a long-term cause might be the case?”
“Could there be an interrelationship of three or four factors that we’re missing?”
What examples do you have?