Spurious correlations: US spending on science and technology VS suicides by hanging

Some time ago we discussed the difference between Correlation and Causality. In short, the fact that two variables (e.g. murders in the US and marketshare of Internet Explorer) evolve in time following a similar pattern (i.e. both murders and the marketshare of IE have decreased in the last years) doesn’t mean that one is a consequence of the other (which means that banning IE would not stop murders, unfortunately). This is obviously very important to keep in mind when optimizing our site for conversions, since we have to make sure that the improvements we see are because of the changes we are testing.

As many things in life, this is easier said than done. We people tend to assume there is a cause-effect in all we do. To combat this bias, this site shows some of the most bizarre correlations you could imagine (e.g. did you know that eating Margarine “affects” the divorce rate in Maine?).

If you don’t optimize in a “scientific” way, what you see can be just pure chance or the effect of some other variable you’re not taking into account. If so, you’re risking taking poor design decisions based on wrong data, which could hurt your business in the future.

Remember, only controlled experiments like A/B testing eliminate the influence of external factors in the result. Start optimizing your site the right way with Nelio A/B Testing today!

Leave a Reply

Your email address will not be published. Required fields are marked *

I have read and agree to the Nelio Software Privacy Policy

Your personal data will be located on SiteGround and will be treated by Nelio Software with the sole purpose of publishing this comment here. The legitimation is carried out through your express consent. Contact us to access, rectify, limit, or delete your data.