Skip to main content
Log in

Summary

The Cornu criterion and the coefficient of skewness test are shown to be preferable to other tests when testing the hypothesis that small samples, less than 100, have been obtained from a normal population. The Cornu ratio of mean deviation to standard deviation is highly negatively correlated with the coefficient of kurtosis.

Zusammenfassung

Es wird gezeigt, daß der Cornu-Test und der Test mittels des Schiefekoeffizienten anderen Tests vorzuziehen sind, wenn man prüft, ob kleine Stichproben mit weniger als 100 Beobachtungen aus einer Normalverteilung stammen. Das von Cornu verwendete Verhältnis zwischen der mittleren Abweichung und der Streuung ist stark negativ mit dem Exzeßkoeffizienten korreliert.

Résumé

On démontre ici que le test de Cornu et celui du coefficient d'obliquité est préférable à tout autre lorsqu'il s'agit d'examiner si de petits échantillons de moins de 100 observations proviennent d'une répartition normale. Le rapport utilisé par Cornu entre l'écart moyen et l'écart quadratique moyen est corélé d'une manière fortement négative avec le coefficient de l'excès.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Brooks, C. E. P., andN. Carruthers: Handbook of Statistical Methods in Meteorology. London, 1953.

  2. Cochran, W. G.: Some Methods for Strengthening the Common χ2 test. Biometrics10, 417–451 (1954).

    Google Scholar 

  3. Geary, R. C.: On the Correlation ofB 2 andW, Section III of the Ratio of the Mean Deviation to the Standard Deviation as a Test of Normality. Biometrika27, 310–355 (1935).

    Google Scholar 

  4. Geary, R. C.: Moments of the Ratio of the Mean Deviation to the Standard Deviation for Normal Samples. Biometrika28, 295–307 (1936).

    Google Scholar 

  5. Pearson, E. S.: A Further Development of Tests for Normality. Biometrika22, 239–249 (1930).

    Google Scholar 

  6. Siegel, S.: Nonparametric Statistics P. 47, New York, 1956.

Download references

Author information

Authors and Affiliations

Authors

Additional information

With 1 Figure

Rights and permissions

Reprints and permissions

About this article

Cite this article

O'Brien, J.J., Griffiths, J.F. Choosing a test of normality for small samples. Arch. Met. Geoph. Biokl. A. 16, 267–272 (1967). https://doi.org/10.1007/BF02246403

Download citation

  • Received:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02246403

Keywords

Navigation