The probability of the outcome's occurrence by "chance" is 5% or less. "Chance" results are those that are within normal variation, which is defined as 2 standard deviations from the mean. Results that are less than 2 standard deviations from the mean are considered not statistically significant, meaning that the null hypothesis should not be rejected. There are some differences the point at which the 5% case occurs, depending on whether the null hypothesis is directed or not.
