Why do we so often use 0.05 for hypothesis testing?
In this online exercise, we hope to help you gain an improved understanding of what a significance level is, and why a value in the neighborhood of 0.05 is reasonable as a default.
Watch this video before starting
We consider two competing hypotheses
- H0: There is no real trend. Any apparent trend is due to chance.
- HA: There is a real, upward trend.
There is a clear, upwards trend. Reject H0.
Uncertain if the trend is real. Do not reject H0.
