Reading a result

Significance, size and uncertainty.

What p-values say

A small p-value means the data would be surprising if nothing were going on. It is not the probability the hypothesis is true, and it says nothing about size.

With a large enough sample, trivial differences become significant.
Confidence intervals communicate far more than a threshold does.

Effect size

Ask how big the effect is in units you care about, not just whether it passed a test. A statistically significant improvement of one percent may be worth nothing.