The newest book to tear through the mathematics and statistics world is the number one bestseller: The Data Detective by Time Harford. In it, we are led on a journey of what it truly means to think more like a scientist, and less like one afraid of numbers. The lessons in the book truly sets strong guidelines for how to make better more informed decisions as a manager or leader, and not fall into the everyday traps of which people tend to not be aware.
- “If we don’t master our emotions, whether they are telling us to doubt or telling us to believe, we’re in danger of fooling ourselves.”
This is truly at the heart of all decision making. Being able to understand emotional response, as opposed to recognizing the emotional bias and removing it from the equation, is key to making decisions based on data and not “gut reaction” or “instinct”. Sometimes instinct is right; sometimes it’s wrong. A true manager understands that instinct can be wrong and does not feel slighted or upset when that occurs.
- “Try to take both perspectives”
Taking both sides of the perspective can help distinguish between truth in data and truth in emotions. Every time you feel like making any decision, try to take the exact opposite position. See what the data says; see what emotional response you have; look at how that argument might change someone’s mind. All of this “devil’s advocate” work helps put fact in check with emotion and leads to a clearer direction on the decision.
- “Ask what is being counted”
When viewing any graph look at exactly what people were asked; look at the numbers; look at how important the issue is in relation to the overall scope of the challenge, opportunity, or problem. A big curve on a chart about a question that plays no role in the customer’s decision to buy or not buy is a common trap. While a survey respondent may say an issue is important, did it play a role in their purchase decision? Interestingly, this is where Bayes’ Law comes in handy.
- “Look for something that will give you a sense of scale”
The sense of scale is closely related to “asking what is being counted.” Too often, the subset of data is un-circumstantial to the overall target population, and in fact may be a group that should purposefully be avoided. The authors hone in on this point saying that “‘n = All’ is often a seductive illusion: it’s easy to make unwarranted assumptions that we have everything that matters. We must always ask who and what is missing.”
- “Pause for a moment to notice how the graph makes you feel: triumphant, defensive, angry, celebratory? Take that feeling into account.”
This is one of the most important take-aways from the book simply because it illustrates the powerful effect of stepping back and asking why exactly am I having this emotional response? Where is it coming from and why? The authors note a famous forecaster who ultimately had a poor track record, not because of his understanding of the numbers but because he truly believed the “world is ruled by figures instead of feelings.” Being cognizant that people make emotional decisions does not mean you have to as well. Separate the two; make better decisions for yourself while realizing that your customer base, vendors, and even colleagues will most likely make the emotional one.