P-value
Usually to test if mean has changed
1. Get z of (new mean - old mean) / (ō / root of n)
2. p = 2 X P(> |z|)
3. If p < 0.05, mean has changed reject Ho
Ways to test mean hypothesis at 95%
Find sample size n if error is y%
Not told standard dev
1/root n = 0.y
What to remember when writing confidence interval at 95%
1.96 X the error