Definition of A-square:
A-squared is the test statistic for the Anderson-Darling Normality test. It is a measure of how closely a dataset follows the normal distribution. The null hypothesis for this test is that the data is normal. So if you get an A-squared that is fairly large, then you will get a small p-value and thus reject the null hypothesis. Small A-squared values imply large p-values, thus you cannot reject the null hypothesis.