Unicorn: A Program for Transforming Data to Approximate Normality

Abstract

Unicorn, a BASIC computer program that automates the Box-Cox transformation, is described. Unicorn reads free field ASCII data files, is completely interactive, and is user friendly. Unicorn can handle files with or without missing data. Descriptive statistics and tests of normality are provided for each variable in untransformed format. Statistics include the first four moments of the distribution, the third and fourth L-moments, the Kolmogorov-Smirnov test, and D'Agostino et al.'s tests of normality. The user is then offered a choice of minimizing skewness or departures from normality overall. The optimal transformation is then found and descriptive statistics and tests of normality are provided for the transformed data. A new ASCII data file is written with transformed data. This new data file is then available for further analysis.

Get full access to this article

View all access options for this article.

References

Alber, S. A. , & Schnaffer, D. W. (1992). Evaluation of data transformations used with the square root and school field models for predicting bacterial growth rate. Applied and Environmental Microbiology, 58, 3337-3342.

Box, G.E.P. , & Cox, D. R. (1964). An analysis of transformations. Journal of the Royal Statistical Society, 26, 211-243.

Berry, W. D. (1993). Understanding regression assumptions. Newbury Park, CA: Sage.

Biscay Lirio, R. , Valdes Sosa, P. A. , Pascual Marqui, R. D. , Jimenez-Sobrino, J. C. , Alverez Amador, A. , & Galan Garcia, L. (1989). Multivariate Box-Cox transformations with applications to neurometric data. Computers in Biology and Medicine, 19, 263-267.

D'Agostino, R. B. , Belanger, A. , & D'Agostino, R. B. (1990). A suggestion for using powerful and informative tests of normality. American Statistician, 44, 316-321.

DeStefano, A. L. , & Van Vleck, L. D. (1991). Heritability estimates of transformations of normally distributed records. Journal of Dairy Science, 74, 1408-1412.

Dunlap, W. P , & Rasmussen, J. L. (1982). A program to determine data transformation maximizing linear regression. Behavior Research Methods and Instrumentation, 14, 357-358.

Edgington, E. S. (1969). Statistical inference: The distribution-free approach. New York: McGraw-Hill.

Glass, G. V , Peckham, P. D. , & Sanders, J. R. (1972). Consequences of failure to meet assumptions underlying the fixed effect analysis of variance and covariance. Review of Educational Research, 42, 237-288.

10.

Hays, W. L. (1988). Statistics (4th ed.). Philadelphia: Holt, Rinehart & Winston.

11.

Hopkins, K. D. , & Weeks, D. L. (1990). Tests for normality and measures of skewness and kurtosis: Their place in research reporting. Educational and Psychological Measurement, 50, 717-729.

12.

Kirk, R. E. (1982). Experimental design. Monterey, CA: Brooks/Cole.

13.

Micceri, T. (1989). The unicorn, the normal curve, and other improbable creatures. Psychological Bulletin, 105, 156-166.

14.

Rasmussen, J. L. (1989). Data transformation, Type I error rate and power. British Journal of Mathematical & Statistical Psychology, 42, 203-213.

15.

Royston, P. (1992). Which measures of skewness and kurtosis are best? Statistics in Medicine, 11, 333-343.

16.

Schork, N. J. , Weder, A. B. , & Schork, M. A. (1990). On the asymmetry of biological frequency distributions. Genetic Epidemiology, 7, 427-446.

17.

Sullivan, L. M. , & D'Agostino, R. B. (1992). Robustness of the t test applied to data distorted from normality by floor effects. Journal of Dental Research, 71, 1938-1943.