Theoretical and Applied Genetics

, Volume 52, Issue 5, pp 201–207 | Cite as

The influence of aberrant values on the statistics related to a selection program

  • B. Thompson


Examples are presented to illustrate some of the effects aberrant values, in particular, measurement errors, may have on estimates of the genetic parameters related to selection studies. It is shown that aberrant values may cause observed response to selection pressure to differ considerably from predicted response. Possible dangers of indiscriminate screening are also discussed.

Key words

Outliers Selection experiment Analysis of variance Heritability Data screening 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Anscombe, F.J.: Rejection of outliers. Technometrics 2, 123–147 (1960)Google Scholar
  2. Finney, D.J.: Problems, data, and inference. J.Roy. Statist. Soc. A 137, 1–23 (1974)Google Scholar
  3. Grubbs, F.E.: Sample criteria for testing outlying observations. Ann. Math. Statist. 21, 27–58 (1950)Google Scholar
  4. Hawkins, D.M.: The detection of errors in multivariate data using principle components. J. Am. Statist. Ass. 69, 340–344 (1974)Google Scholar
  5. Kruskal, W.H.: Some remarks on wild observations. Technometrics 2, 1–3 (1960)Google Scholar
  6. Prescott, P.: An approximate test for outliers in linear regression. Technometrics 17, 129–132 (1975)Google Scholar
  7. Thompson, B.K.: Some statistical problems in the analysis of poultry data. Ph.D. Thesis. University of Edinburgh (1974)Google Scholar

Copyright information

© Springer-Verlag 1978

Authors and Affiliations

  • B. Thompson
    • 1
  1. 1.Statistical Research ServiceResearch Branch Agriculture CanadaOttawaCanada

Personalised recommendations