How data analysts experience cognitive biases and how to recognise them: Part 3

29 March 2018

Cognitive biases: We all have them and we regularly encounter them in everyday life. But how do data analysts actually experience these biases? And how can you recognise them? In this series we will discuss cognitive biases in the context of analysts. Each blog picks out 3 biases from 20 categories.

In this blog, we have three biases from the “Not Enough Meaning” spectrum: Essentialism, the Positivity Effect and the Appeal to Probability Fallacy.

Essentialism

Essentialism means that you believe that every group has certain characteristics that are relevant to the identity of that group. That sounds terribly vague, but any analyst who does cluster analyses will certainly recognise this. In cluster analyses you often find a cluster for which you are unable to find any genuinely distinctive characteristics. Sometimes you end up looking for a variable that is distinctive to that specific cluster: surely these customers MUST have something that distinguishes them from others. And so finally you end up (by coincidence?) with a set of variables that are indeed different, and so you define your cluster as “Women Who Live in the Netherlands and Drink Mint Tea”… but whether that is in fact relevant to that group remains to be seen…

Positivity Effect

People focus more on positive information and remember it better: The Positivity Effect. And the older you get, the stronger this bias becomes. So when you tell your colleagues, particularly if they are a little older, that you’ve got some good news (“we can allocate customers to segments for sending the newsletter”) and some bad news (“the model has a hit rate of just 58%”), it is likely that they will take the good news on-board and forget the rest. They can’t help it, it’s natural.

Appeal to Probability Fallacy

When people think something is probably going to happen, they automatically assume that it will actually happen. Or if people think something probably won’t happen, they assume that it isn’t going to happen. In other words, the Appeal to Probability Fallacy says that you turn high probabilities into 100% probability and low probabilities into 0% probability. This can come in very handy: if you say that there is an “extremely low probability” of you finishing your analysis this week, the listener will assume that it isn’t going to happen this week. If when it comes to Friday afternoon you do manage to finish your analysis after all, you score bonus points! But watch out, by the same token, if you say “I’ll probably be able to find a significant model”, the listener will assume you are definitely going to find a suitable model!

Also read the other parts:

Part 1: Self-relevance effect, Confirmation bias and the Bias blind spot.

Part 2: Availability heuristic, Contrast effect and Clustering illusion.

Part 4: Curse of Knowledge, the Hindsight Bias and the Hard-Easy Effect.

Part 5: Sunk Cost Effect, the IKEA effect and the Google Effect.

Part 6: Ambiguity Bias, Appeal to Novelty and False-Consensus Effect.

Contact

Do you want to know more about this subject? Please contact us on +31 (0)33 258 28 30 or info@cmotions.nl.