The Problem With Statistical Evidence

While statistics are extremely valuable, they are also notorious for being a means that people use to make false and misleading arguments.

86% of statistics are made up on the spot, you know – the remaining 24% are mathematically flawed.- Anonymous

Also far too often statistics are obtained by taking a sample from a larger group and assuming the whole group has the same characteristics as the sample. Which is just bad science, rather than straight out lying.

Then there’s the problem with statistics that are true but misleading

Even when statistics are technically accurate, particular statistical facts can be very misleading. So statistics can certainly be useful, but should never be trusted as evidence. Statistics when used to prove something always need to be accompanied by information on how it was gathered.

Here are some excellent, and rather humorous examples of just that:

These graphs were made using software that finds spurious correlations, and some of them are quite entertaining to say the least.

So when presented with a claim accompanied by statistical evidence, remember to stay skeptical. Question the methods of how they were gathered as well as the circumstances. Statistics should not be enough to convince an intelligent human being.

Comments

comments