TEORlA y APLlCACION DE LA TRANSFORMACION DE VARIABLES EN GEOGRAFIA

ALDEN  GAETE JENICEK

Authors

ALDEN GAETE JENICEK Pontificia Universidad Católica de Chile

Keywords:

Analysis of variance, geographical theory, applied geography

Abstract

The validity of conclusions drawn from an Analysis of Variance depends upon the extent to which the assumptions of the technique are satisfied by the data. The assumptions are those of normality randomness and independence, additivity and homoscedasticity. The present study contains three basic sections. The first demostrates that a situation where the assumptions are not met may be remedied by the creation of a new variable obtained by transforming the original data. In the second, the reasons which may cause the data to differ from the assumptions are discussed in conjuntion with descriptions of several appropriate transformations. Finally, the third section presents some theoretical and practical considerations arlsing from data transformation, also mentioning some nonparametric techniques which can be used instead of the usual transformations.

References

BARTLETT, M. S., 1936.-"The Square-Root Transformation in the Analysis of Variance". Suppl. Jour. Roy. Stat. Soc., 3, 38-78.
BARTLETT, M. S., 1937.-uSome examples of Statistical Methods of Research in Agriculture and applied Biology." Suppl. Jour. Roy. Stat. Soc., 4, 137-183.
BARTLETT, M. S., 1947.-uThe Use of Transformation". Biometrics, Vol. 3, 39 p.
BARTLETT, M. S. and KENDALL, D. G., 1946.-"The Statistical Analysis and the Logaritmic Transformation". Suppl. Jour. Roy. Stat. Soc., 8, 128-138.
BEALL, G., 1942.-uThe Transformation from entomological Field Experiments so that the Analysis of Variance becomes applicable". Biometrika, 32, 243-262.
BENNET, A. C. and FRANKLIN, L. N., 1954.-Statistical Analysis in Chemistry and the Chemical Industry. New York, John Wiley.
BLISS, 1. C. and CALHOUN, W. D., 1954.-An Outline 01 Biometry. New Haven, Connecticut, Cooperative Corporation.
BROWNLEE, A. K., 1966.-Statistical Theory and Methodology, 2 ed. New York, John Wiley.
COCHRAN, W. G., 1947.-"Some Consecuences when the Assumptions for the analysis of Variance are not satisfied". Biometrics. Vol. 3, pp. 22.
DAVIS, J. T., 1971.-"Sources oí Variation in Housing Values in Washington, D.C." Geographical Analysis. 3:63-76.
DIXON, J. N. Y MASSEY, J. F., 1970.-Introducción al análisis estadístico. 2 ed. México, Mc. Graw-Hill.
BEDERER, T. W., 1955.-Experimental Design Theory and Application. New York, The Mac. Millan.
FISHER, R. A., 1950.-Statistical Methods, lor Research Workers, 11th, ed. Edimburgo Oliver and Boyd.
HAGGET, P., 1967.-Locational Analysis in Human Geography, 3 ed. London, Edward Amold.
KEMPTHORNE, O., 1967.-The Design and Analysis 01 Experiments. 6 Ed. New York, John Wiley.
KING, J. L., 1969.-Statistical Analysis in Geography, Englewood Cliíís, N. Y. Prentice- Hall.
KNOSS, D. S., 1962.-Distribution 01 Land Values in Topeka, Kansas. University oí Kansas, Bureau oí Business and Economic Research.
KRUMBEIN, W. C. and MILLER, R. L., 1954.-"A Note on Transíormation oí Data íor Analysis oí Variance". Journal 01 Geology, Vol. 62, 1954, p. 192-193.
LI, C. C., 1969.-Introducción a la estadística experimental, Barcelona, Omega.
MAXWELL, C. J., 1967.-Quantitative Geomorphology of some Mountains Chaparral Watersheds of Southern California. DEPARTMENT OF GEOGRAPHY, North
Western University, Evanston, Illinois.
MURDIE, R. A., 1969.-F'actorial Ecology of Metropolitan Toronto 1951-1961, An Essay on the Social Geography of the City. (Research paper Nr. 116). Chicago:
Department oí Geography, University oí Chicago, 212 P.
SIEGEL, S., 1970.-Diseño experimental no paramétrico aplicado a las ciencias de la conducta. México, Trillas.
SNEDECOR, W. G., 1964.-Métodos estadísticos aplicados a la investigación agrícola y biológica. 4 ed., México, Continental.