Confirmation Bias

Turkis Stars at BIAS 2018 in Romania
Photo by Tudose Alexandru / Unsplash

I’m not immune to confirmation bias. And if someone suspects that I’m perpetuating an opinion that might not be the whole story, I’d like for them to stop me and remind me of that as a starting point for debate.

Present me some conflicting data.

I’ve spent a great deal of time lately looking at fitness, nutrition and psychological research. I try to read with an open mind, but alas, I often stumble across information that confirms my beliefs rather than opposes my beliefs.

When I do stumble across information that opposes me beliefs, I am probably far more likely to be critical of it as well.

It’s important to be skeptical, even of information that supports your beliefs.

Were there potential research flaws? Is there something from your previous experience that might rule out the applicability of certain pieces of research.

For instance, I’m big on dynamic warm-ups and not static stretching before training. Static stretching has been shown to decrease power output for about 20 minutes post-stretch, leading to decreases in performance.

HOWEVER… if you have time to stretch more than 20 minutes before performance and a person needs improved flexibility, it’s still useful. If I have someone who benefits from it, and I have more than 20 minutes to game time, why not use it still?

Context Matters!

The internet ties information together that it thinks will be of similar interest to you. Deep learning networks are tying context together for you in the background all the time.

Google and Facebook recommend similar links based on the things you’ve clicked on. Even PubMed has a section on their website called, “Related Citations” to encourage you to read other abstracts or papers that are similar to what you’re currently reading based on their algorithms.

I often click through in the hopes of finding something similar but different. I want the whole story and not just a potentially random piece of research that seemingly provides a black and white answer. Most answers to questions are grey.

The two thing that keeps me going in respect to this are:

  1. Crafting opinions publicly, ala blogging, encourages debate, so I often have people willing to debate and show me a little of the other side of the coin.
  2. Information truly exists on a spectrum.

1 is fairly obvious, I’m here blogging about my experiences and the comments are open if you want to present something conflicting or supportive.

I think the smartest people around change their minds when necessary. I’d like to think that if presented with a convincing enough argument you could probably change my mind on at least one small thing within the debate.

This has given rise to me changing the wording on many of my posts over the years as my views change ever so slightly. This is even more pressingly true on Quora where I have some 700 answers to questions now and almost one million views.

B is a more interesting self-discovery…

Thinking of information on a spectrum has dramatically changed my outlook on research and practicing my craft.

I very rarely view anything any more as either good or bad, but rather, my choice of usage is influenced by where it falls on a spectrum of choice, based on context.

<-Worse———————Better->

I’m using a spectrum of choice to make decisions about where to go in coaching, based on perceived need.

For instance I know that exercising 6 days a week is better than 2, I know that, you know that. But if a person is exercising zero days, then one day a week is a noticeable improvement.

I don’t start with the ‘best strategy’ if it doesn’t apply to the situation contextually, and am trying to remain contextually applicable, rather than biased to right and wrong.

I know that squatting is a great exercise for most, but every now and then I still come across someone where it might not be a good idea.

You can’t fit a square peg in a round hole.

No matter how smart or intelligent you might think you are, you’re not immune to confirmation bias. I’m not a genius but even if I was, I wouldn’t be immune to confirmation bias any more than the next person.

And it’s seriously rampant in the world, even among researchers. Just because you have experience with research, also doesn’t mean you don’t suffer from unconscious bias. It’s a large reasons why the double blind study has become the gold standard in research, because in theory you can remove the research bias.

Though the researcher can still publish conclusions based a double blind study, and thus adjust stats quite easily or make outlandish claims by manipulating data.

And this still happens, but it’s a completely separate issue.

I think it’s important to stop and check yourself regularly, before you wreck yourself:

  • Are your views on this subject too absolute (black and white) to be useful? And most likely incorrect…
  • Have you read research from both sides of the argument? Or at least tried?
  • Does the conflicting research present interesting data on both sides of the equation that leads the the likely conclusion that the answer lies somewhere in the middle?
  • Does your information come from more than one source? Ideally more than 3 different people at a minimum, but 10 or more would be better if it exists…
  • When you present your information, do you do so with accounting for conflicting information?
  • Can you rationally explain why conflicting data exists and if it’s true, would it change your mind?

It’s easy to spot a guru, or bullshit, if you stop for a moment and ask yourself a few of these questions. If a person conveniently ignores any conflicting data, they are likely deeply entrenched in confirmation bias.

A person who openly presents the conflicting views and draws a conclusion somewhere in the middle based on their best judgement through data they’ve seen. Is, in my mind, probably worth listening to.

If they can budge on their argument, when new data is presented that offer a slightly different point of view, that person is probably more trustworthy too. A person who is willing to change their method, based on the situation, is confirming the method with a result. At least that’s something…

One of my favourite strength coaches (Mike Boyle) once said, “don’t believe everything you read and don’t read everything you believe.

That last bit is important…

Darren Beattie

Read more posts by this author.