The Null Hypothesis - H0: The random errors will be normally distributed
The Alternative Hypothesis - H1: The random errors are not normally distributed
Reject H0: when P-value ≤ α = 0.05
As the P value is 0.005 it is less than the 0.05 significance level therefore reject H0 and accept H1 as there is sufficient evidence to suggest that random errors are not normally distributed. The assumption of normality is possibly satisfied as the normal probability plot is close to the straight line.