So the US presidential polls turned out… shall we say not entirely right? And it isn’t the first time this year: think of Brexit or the Colombian referendum. Something’s gone awry in the numbers.

Numbers are pretty cool because they provide objective, factual answers. In science, you ask a question, then go measure the answer, which will come in the form of a number. That number is a fact, which you can use to prove other facts.

Numbers are also tricky, because they answer without any comment on the question. Was it posed correctly? Was it stupid? Was it the right question to begin with?

If we pretend we asked something we didn’t, or try to read in the numbers something that we didn’t ask for, the whole point is moot. And that’s when statistics look arbitrary.

But that’s not the numbers’ fault. It’s about who reads them and how they read them. It’s about the question.

This reminds me of The Hitchhiker’s Guide to the Galaxy. A super-intelligent alien race builds Deep Thought, a fenomenal supercomputer tasked to find “the answer to life, the universe, and everything”. After millions of years of calculations, the answer comes (in the form of a number, no less), but—mild spoiler—it’s kind of underwhelming.

As you can see, the question was just as important as the answer, no matter how true (and my heart has no doubt that is the answer), accurate or objective. And you don’t get much more objective than an accurately measured numerical value.

Here’s, I think, the thing about polls and elections: pollsters know what they are doing. They know how to account for all possible biases and how to accurately measure something that always proved to reflect voting reults. But they cannot know whether the vote is going to go that way. That’s not the question.

So what’s wrong? We don’t know, but hordes of statisticians are surely hellbent on figuring it out.

The awesome power of numbers is to tell us exactly what we asked for or, if the answer makes no sense, show we asked it wrong. The awesome skill of scientists, their art even, is to find the right questions and the right way to ask them.

Statisticians will figure it out, and we’ll learn a cool lesson from it. Silver linings, people.

Cover photo: CC0 Andrew Martin, via pixabay


Questions? Comments? Let's talk about that!

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s