Some of my variable categories are being shown to have y-values higher or lower than the overall average. Why aren’t they all identified as statistically significant?
Ultimately, the tests of statistical significance rely on the evidence contained in the dataset. If a variable is not found to be significant, it could be due to a lack of sufficient data (not enough records/values). It may also indicate, more generally, that the variable does not effectively explain a difference in the predicted outcome, historically.