The critical value of K-S statistic at 90%, 95%, or 99% significance level are indicators of normal distributions. For example, if a sample collected from a population has a normal frequency distribution, the K-S statistic for that sample is less than the critical value 90, 95, or 99 percent of the time. If the K-S statistic is larger than the critical value, the hypothesis that the underlying population is distributed normally with a mean of and a standard deviation of s should be rejected.
See Also
95% and 99% Confidence Interval for the Mean
Kolmogorov-Smirnov Goodness of Fit Statistic for Normal Distribution