ARTICLES

# Learning about p-value

What do people want you to think whey they say “p-value”? The top hit from Google says:

The p value is the evidence against a null hypothesis. The smaller the p-value, the stronger the evidence that you should reject the null hypothesis. P values are expressed as decimals although it may be easier to understand what they are if you convert them to a percentage. For example, a p value of 0.0254 is 2.54%.

I’m not sure what to make of those statements. A beetle may be stronger than an ant, but what I really care about when
I’m hiking with my family is whether a creature poses a real threat to my kids. The magnitude and implication of
*stronger* matters. This is why we use the Richter scale when talking about earthquakes instead of using absolute
numbers.

Expressing p-value as a percentage also falls short. 2.54% of… what? Take a hypothesis “proven” by a study with a p-value of 0.0254, say (this is made up), the hypothesis that men with shoulder-length hair are more lkely to be gay. Does that mean exactly 2.54% of men with shoulder-length hair are not gay? Does that mean given the same man with shoulder-length hair, if you asked him if he is gay sufficient times, then he will say “no” 2.54% of the time and “yes” 97.46% of the time? In both cases, we “converted” 0.0254 to a percentage, but obviously, both interpretations are ridiculous.