Comparison of Normality Tests in Terms of Type-I Error and Power with Different Sample Sizes and Distributions

Authors

  • Mustafa Cagatay Buyukuysal Zonguldak Bulent Ecevit University, School of Medicine, Department of Biostatistics, Zonguldak, Turkey
  • Vildan Sumbuloglu Sanko University, School of Medicine, Department of Biostatistics, Gaziantep, Turkey

Keywords:

Normal distribution, Normality tests, Type-I error, Power

Abstract

The aim of the study was to compare the power and Type-I error rates of 5 different normality tests with different distributions and sample sizes. For each sample sizes (n=7, 10, 30, 50, 100), 500,000 data sets generated from a normal distribution to test the Type-I error rates of Anderson-Darling, Cramer-von-Mises, Jarque-Bera, Kolmogorov-Smirnov and Shapiro-Wilk tests. Type-I errors of the tests were calculated by computing the rejected null hypothesis over 500,000 samples for each distribution. To evaluate
the power of the tests, Chi-square (

Downloads

Published

15-09-2021

How to Cite

Cagatay Buyukuysal, M., & Sumbuloglu, V. (2021). Comparison of Normality Tests in Terms of Type-I Error and Power with Different Sample Sizes and Distributions. International Journal of Basic and Clinical Studies, 10(2), 57–65. Retrieved from https://www.ijbcs.com/ijbcs/article/view/ijbcs10205

Issue

Vol. 10 No. 2 (2021)

Section

Original Article

License

Copyright (c) 2021 by the Authors

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.