Comparison of Normality Tests in Terms of Sample Sizes under Different Skewness and Kurtosis Coefficients

This study aims to compare normality tests in different sample sizes data with normal distribution under kurtosis and skewness coefficients obtained simulatively. To this end, firstly, simulative were produced using the MATLAB program for skewness/kurtosis sizes. The analysis of each type was conducted ten tests; namely, (Kolmogorov Smirnov (KS) Test, KS Stephens Modification, Marsaglia, Lilliefors Anderson-Darling Cramer- Von Mises Shapiro-Wilk Shapiro-Francia Jarque-Bera D’Agostino & Pearson Test). As a result analyses according tests, it found that not affected by size when equal or close zero; however, cases where moved away from zero, are size, such tend give significant results. Therefore, large samples, may be suggested critical values coefficients’ z-scores as proposed Kim (2013) Mayers histogram graphs used.