Selection Bias

It’s time to learn why you can’t trust anything on the internet.

Today we’re going to dive into the world of statistics and have a look at Selection Bias. And because we’re talking about the internet, we’re going to demonstrate the principle of selection bias by looking at Blaxploitation films on IMDB.

I’ve listed below the ratings and vote counts of four marvelous examples of the genre.

Film IMDB Rating # of votes
Black Belt Jones 6.0 805
Hot Potato 4.3 81
Three the Hard Way 5.8 246
Black Samson 5.7 94

Looking at these numbers, they don’t look horrible, all things considered. The most popular, Black Belt Jones, weighs the scales at a respectable 6.0 with 805. Looks good, huh?

There’s only one problem. Black Belt Jones is *not* a 6.0 film. Trust me, folks (and don’t ask me how I know).

What this 6.0 *really* means is not that the film is a 6.0 film. It means that the dedicated fans of the genre who took the trouble to hunt down this difficult to find, niche film, and who cared enough to vote gave it a six. The people voting on this film are people who are predisposed to like it.

The numbers only get worse from here – 5.8, 5.7, 4.3 – but the really gruesome numbers are the vote counts. Trust me, you don’t want to take the word of those 94 people… they are the wrong 94 people to listen to about this film.

That, in a nutshell, is selection bias. This is why you can’t trust anything on the internet, because *everything* on the internet is heavily subject to selection bias. Well, you can, but…

It means you might wind up watching Black Samson one night.

This entry was posted in tongueincheek and tagged , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published.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

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>