Data: A Spotter’s Guide to Fallacies

Our minds are liars. Not just in the sense that we routinely think things we know not to be true to assuage our guilt for doing something, such as “he probably didn’t want this one anyway” when you steal a chip off your friend’s plate while he’s at the bar. That’s a simple untruth, consciously thought up as an excuse. But your mind is a more insidious liar than that – it deceives you on a deeper level, by making you make incorrect judgements based on erroneous, false patterns.

In Paranormality, Professor Richard Wiseman explains it this way:

Let’s imagine you are out in the wilderness and the wind causes some nearby bushes to rustle. You have been told there are several hungry tigers in the area and know they cause the same type of rustling sound. [] In terms of your long-term survival, you are better to err on the side of the tiger hypothesis. To put it in more psychological terms, it is better to see a few patterns that are not there than miss one that is.

Because of this, your pattern-finding skills have a built-in tendancy to find connections between completely unrelated events.

Tigers in the mind

Tigers in the mind

A fun and ubiquitous demonstration of this effect is the phenomenon of pareidolia; the tendency of the mind to see faces where there are none. Jesus in a bagel, or the man onsurface of the moon, or similar. Because the ability to recognise faces is so deeply ingrained, we can’t not see them. Pareidolia is a type of apophenia, seeing patterns in random data. It’s also known as a Type I Error because it involves falsely finding a connection between two unrelated pieces of information.

Part of the reason apophenia is such a danger when it comes to analysing data is because the human mind is also hopelessly prone to confirmation bias; the tendency to give more weight to that which confirms a pre-existing belief. In Irrationality, Stuart Sutherland explains:

[] people try to prove that their current hypothesis is correct – they test it only picking only examples that will confirm it and do not look for ones that would disconfirm it.

As evidence, he cites this study by Peter Wason. Of all the habits to which the mind adheres, confirmation bias is the ultimate scourge of rational thinking. Combined with our tendency to apophenia, it means we’re naturally inclined to misinterpret or wilfully ignore data. Dr Ben Goldacre, author of the fantastic book on this topic Bad Science, provides this horrifying example.

So, what can we do to prevent ourselves misinterpreting data, through our pattern-recognition or confirmation bias? Part of the solution is simply to be aware of our tendencies. Once you know the fallacies that we routinely fall into, it is easier to avoid them. It’s not a foolproof method, since fallacies such as the ones mentioned above are so insidious that we fall prey to them constantly, but it’s a start.

Related reading:

Tagged with: , , , , , , ,
Posted in Data, Thinking

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: