How to lie with true statistics

This graph really, really annoyed me:

I’ve seen this graph a dozen times; this one is from politico magazine, but that doesn’t matter: It’s horribly misleading.

It is about the most dishonest graph I’ve ever seen – and I’ve seen plenty.

Why?

Well – do you notice that arbitrary line at “2/3 vote” drawn into the graph? Makes the “yes” vote look really wanting to extend to that line, doesn’t it?

Sure, the 2/3rds majority is necessary, but that’s a democratic vote result – usually shown in block graphs:

Oh, look; I’ve managed to draw exactly the same graph with exactly the same arbitrary line at 2/3rds, but now the prospector might actually get the really important information? Hint: Guess which vote got the majority?

Which, ignoring the arbitrary line, should be – well, you can see that in my graph. Just like if the function of media were to report truth.

See- I can skew that close a result without being that dishonest….

Still the same graph, you see? No? Learn fucking statistics or at least:

Please pay attention to the numbers before looking at a graph. You can make any graph quite arbitrarily “say” anything you want – e.g., by inserting random lines, even if you are stupid. True does not exclude misleading.

Oh, btw, that makes you a demagogue.

One Reply to “How to lie with true statistics”

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.