The Use of the Arc-Sine Transformation in the Analysis of Variance
- 1 September 1987
- journal article
- Published by SAGE Publications in Educational and Psychological Measurement
- Vol. 47 (3) , 563-573
- https://doi.org/10.1177/001316448704700303
Abstract
The use of the arc-sine transformation in the analysis of variance is discussed. The application of the transformation can lead to difficult inference situations and pose interpretational problems for the researcher. An example data set is used to illustrate the difficulty. Additional Monte Carlo results are presented which suggest that the application of the transformation has no effect on the Type I error rate. Furthermore, the transformation can produce tests of noticeably lower power when the null hypothesis is false. As such, the transformation is not recommended as a standard data analytical tool.Keywords
This publication has 14 references indexed in Scilit:
- An Algorithm for Testing the Robustness Properties of Two-Way Nonorthogonal Analysis of VarianceEducational and Psychological Measurement, 1985
- Variance Stabilizing Transformations and the Power of the F TestJournal of Educational Statistics, 1981
- An Examination of the Effect of Six Types of Error Perturbation on Fifteen Clustering AlgorithmsPsychometrika, 1980
- Linear Statistical Inference and its ApplicationsPublished by Wiley ,1973
- Consequences of Failure to Meet Assumptions Underlying the Fixed Effects Analyses of Variance and CovarianceReview of Educational Research, 1972
- Power of the Analysis of Variance of Independent Groups on Non-Normal and Normally Transformed DataEducational and Psychological Measurement, 1966
- NON-NORMALITY AND TESTS ON VARIANCESBiometrika, 1953
- Numerical transformations in the analysis of experimental data.Psychological Bulletin, 1949
- THE TRANSFORMATION OF POISSON, BINOMIAL AND NEGATIVE-BINOMIAL DATABiometrika, 1948
- The Use of TransformationsPublished by JSTOR ,1947