Power comparisons of Shapiro-Wilk , Kolmogorov-Smirnov , Lilliefors and Anderson-Darling tests

By Nornadiah binti Mohd Razali, Yap Bee Wah. 2011

4,000 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/dcdc0a0be7d65257c4e6a9117f69e246fb227423

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Variations of Q–Q Plots: The Power of Our Eyes! (2015, background): “The SW test has been shown to be the most powerful in assessing non-normality (Stephens, 1974; Razali and Wah, 2011).”
Validity and Reliability of the Connor-Davidson Resilience Scale (CD-RISC) in Competitive Sport (2016, background): “However, with a sample size over 200, ShapiroeWilks tests are extremely powerful and small deviations from normality can result in significant test results (Ghasemi & Zahediasl, 2012; Razali & Wah, 2011).”
Predicting social media performance metrics and evaluation of the impact on brand building: A data mining approach (2016, methodology): “We adopted the Shapiro–Wilk test to assess if each of the output columns for the features to be predicted followed a normal distribution (Razali and Wah, 2011).”
Computation of Probability Associated with Anderson–Darling Statistic (2018, methodology): “In a Monte Carlo simulation study conducted on the normal distribution, Kolmogorov–Smirnov test has been identified as the least powerful test, while opposite Shapiro–Wilks test was identified as the most powerful test [34].”
Engineering stress culture: Relationships among mental health, engineering identity, and sense of inclusion (2021, methodology): “To provide justification for statistical testing procedures, we calculated skewness and kurtosis and used the Shapiro–Wilk test to determine whether our data are normally distributed (Razali & Wah, 2011).”
Shapiro–Francia test compared to other normality test using expected p-value (2015, background): “Also, we observed overall that the KS test was the worst test for detecting non-normality in data for all the distributions considered in the study, a finding that was corroborated by others.[1,2] To the best of our knowledge, we did not find any study that have demonstrated the importance of the SF”
An empirical analysis of the factors affecting social entrepreneurial intentions (2017, methodology): “If the Sig. the value of Shapiro-Wilk test is greater than 0.05 than data is considered as normally distributed and if it is lesser than 0.05, data is not considered as the normally distributed (Razali and Wah 2011).”
Person–Organization Fit and Turnover Intention: Exploring the Mediating Role of Employee Followership and Job Satisfaction Through Conservation of Resources Theory (2016, methodology): “Two formal tests of normality—Shapiro–Wilk and Kolmogorov–Smirnov—however, indicated that the distribution of the observed values of the dependent variable and the independent variables generally conformed to the normality assumption of regression analysis (Razali & Wah, 2011).”
Difficult situations and ways of coping with them in the experiences of parents homeschooling their children during the COVID-19 pandemic in Poland (2020, background): “It is very powerful and is considered the best test to check the normality of the distribution of a random variable (Razali and Yap 2011).”
Employee adaptive performance and job satisfaction during organizational crisis: the role of self-leadership (2018, methodology): “To test if the variables under study had normal distribution, the Shapiro–Wilk test was performed (Razali & Wah, 2011).”
Maximum entropy-Gumbel-Hougaard copula method for simulation of monthly streamflow in Xiangxi river, China (2015, background): “exceeds the 1-a quantile, then we reject H0 at the level of significance, a (Razali and Wah 2011).”
RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning (2022, methodology): “The normality of the distribution of Dice scores generated by RadImageNet and ImageNet models was confirmed by the Shapiro test (31,32).”
A Statistical Analysis of Summarization Evaluation Metrics Using Resampling Methods (2021, methodology): “To understand if the normality assumption holds for summarization data we ran the Shapiro-Wilk test for normality (Shapiro and Wilk, 1965), which was reported to have the highest power out of several alternatives (Razali and Wah, 2011; Dror et al., 2018, 2020).”
The role of emotional intelligence and self-efficacy on social entrepreneurial attitudes and social entrepreneurial intentions (2017, background): “05, data are not considered as the normally distributed (Razali and Wah, 2011).”