This probability value (or p-value) is related to a test statistic that compares the shape of the sample distribution to the shape of the normal distribution. The p-value, which can range from zero to one, describes how likely it is that the two distributions can be considered the same.
A p-value close to zero (less than a typical cut-off value such as 0.10) provides evidence against the idea that the sample distribution can be considered normal. Other methods of checking for normality include looking at the sample histogram, seeing whether the skewness and kurtosis are close to zero, and examining the normal probability plot.