I know that if a P value is less than 0.05 we can reject the null hypothesis and state that the finding may be statistically significant however I remember reading something about if the 95% confidence interval includes 0 that its not a significant finding or something like that (i tried to find what i was reading but couldnt find it again).
Im just wonder, if we have a p value below 0.05 but have a confidence interval including 0, how do we interpret this? Do we say that it is statistically significant but likely being caused by other factors?
Best Answer
It means that you took a confidence interval with more than 95% confidence. This is technically possible, but misleading to take the confidence interval at another $\alpha$ level than your test.
Existing software packages will give you p-values and CIs for the same $\alpha$ unless you force them to do something else.