Title: 2015 ACVIM Forum Research Abstract Program Document date: 2015_5_27
ID: 3pnuj5ru_728
Snippet: Although statistically significant differences in accuracy were detected among statistical methods, most of them were not considered clinically relevant. The only relevant inaccuracies were found when parametric or robust methods were used on untransformed data with non-Gaussian population distributions. Specificity of the Shapiro-Wilk test to predict population Gaussianity declined as the sample size decreased (92%, 78%, 67%, and 45% for n = 250.....
Document: Although statistically significant differences in accuracy were detected among statistical methods, most of them were not considered clinically relevant. The only relevant inaccuracies were found when parametric or robust methods were used on untransformed data with non-Gaussian population distributions. Specificity of the Shapiro-Wilk test to predict population Gaussianity declined as the sample size decreased (92%, 78%, 67%, and 45% for n = 250, n = 120, n = 60, and n = 30 respectively).
Search related documents:
Co phrase search for related documents- population distribution and sample size: 1, 2
- population distribution and statistical method: 1, 2
- sample size and Shapiro Wilk test: 1
- sample size and statistical method: 1, 2, 3, 4, 5, 6
- sample size and statistically significant difference: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32
- Shapiro Wilk test and statistically significant difference: 1
Co phrase search for related documents, hyperlinks ordered by date