Research Article
BibTex RIS Cite

Ortanca Değerin Kullanılmasının ANOVA F Testi Bakımından Gerçekleşen 1. Tip Hata Olasılığı Üzerine Etkisi

Year 2004, Volume: 10 Issue: 01, 20 - 23, 01.01.2004
https://doi.org/10.1501/Tarimbil_0000000863

Abstract

Bu çal ış mada, normal olmayan populasyonlarda ortalama yerine ortanca de ğ erin kullan ı lmas ı n ı n Anova F testi bak ı m ı ndan gerçekleşen I.Tip hata olas ı l ığı üzerine etkisi ara şt ı r ı lm ış t ı r. Yap ı lan 50000 simülasyon denemesi sonunda, populasyon varyanslar ı homojen iken ortalaman ı n kullan ı lmas ı daha güvenilir sonuçlar verirken, populasyon varyanslar ı heterojen iken populasyon varyanslar ile örnek hacimleri aras ı nda doğ ru bir e ş leştirmenin yap ı lmas ı durumunda ise ortanca değerin kullan ı lmas ı n ı n daha güvenilir sonuçlar verdi ğ i görülmü ştür. Ters e ş leştirmenin yap ı lmas ı durumunda ise her iki istatisti ğ in kullan ı lmas ı durumunda da oldukça sapma!' sonuçlar ı n elde edildiğ i görülmü ştür.

References

  • Alexander, R. A. and D. M. Govern, 1994. A new and simpler aproximation for ANOVA under variance heterogeneity. J. of Educational Statistics, 19, 91-101.
  • Anonymous, 1994. FORTRAN Subroutines for Mathematical Applications. IMSL MATH/LIBRARY. Vol.1-2. Visual Numerics, Inc., Houston, USA.
  • Brown, M. B. and A. B. Forsythe, 1974. The small sample behavior of some statistics which test the equality of several means. Technometrics, 16, 129-132.
  • Krutchkoff, R. G. 1988. One-way fixed effects analysis of variance when the error variances may be unequal. J. Statist. Comput. Simul.,30, 259-71
  • Sokal, R. R. and F. J. Rohlf, 1995. Biometry. The Principles and Practice of Statistics in Biological Research. Third Ed. W. H. Freeman and Co., 887s., New York.
  • Steel, R. G. D., J. H. Torrie and D. A. Dickey, 1997. Principles and Procedures of Statistics. A Biometrical Approach. McGraw-Hill, 666s., New York. USA.
  • Winer, B. J., D. R. Brown and K. M. Michels, 1991. Statistical Principles in Experimental Design. Third Ed., McGraw-Hill, Inc.,1057, USA.
  • Wludyka, P. S. and P. R. Nelson, 1997. An analysis of means type test for variances from normal populations. Technometrics, 39, 274-286.
  • Zar, J. H. 1999. Biostatistical Analysis. Fourth Ed., Prentice-Hall, Inc.,663s., USA.

Effects of Using Median on Type I Error Rates in Terms of ANOVA F Test

Year 2004, Volume: 10 Issue: 01, 20 - 23, 01.01.2004
https://doi.org/10.1501/Tarimbil_0000000863

Abstract

İn this study, effects of using sample median instead of sample mean on Type I error rates, which were respected to Anova F test under nonnormal distributions, was investigated. At the end of 50,000-simulation trials it was seen that using the mean provided more reliable results than using median. On the other hand, when group variances were heterogeneous and direct pairing was applied using the median provided reliable results. When inverse pairing was applied both tests more deviants.

References

  • Alexander, R. A. and D. M. Govern, 1994. A new and simpler aproximation for ANOVA under variance heterogeneity. J. of Educational Statistics, 19, 91-101.
  • Anonymous, 1994. FORTRAN Subroutines for Mathematical Applications. IMSL MATH/LIBRARY. Vol.1-2. Visual Numerics, Inc., Houston, USA.
  • Brown, M. B. and A. B. Forsythe, 1974. The small sample behavior of some statistics which test the equality of several means. Technometrics, 16, 129-132.
  • Krutchkoff, R. G. 1988. One-way fixed effects analysis of variance when the error variances may be unequal. J. Statist. Comput. Simul.,30, 259-71
  • Sokal, R. R. and F. J. Rohlf, 1995. Biometry. The Principles and Practice of Statistics in Biological Research. Third Ed. W. H. Freeman and Co., 887s., New York.
  • Steel, R. G. D., J. H. Torrie and D. A. Dickey, 1997. Principles and Procedures of Statistics. A Biometrical Approach. McGraw-Hill, 666s., New York. USA.
  • Winer, B. J., D. R. Brown and K. M. Michels, 1991. Statistical Principles in Experimental Design. Third Ed., McGraw-Hill, Inc.,1057, USA.
  • Wludyka, P. S. and P. R. Nelson, 1997. An analysis of means type test for variances from normal populations. Technometrics, 39, 274-286.
  • Zar, J. H. 1999. Biostatistical Analysis. Fourth Ed., Prentice-Hall, Inc.,663s., USA.
There are 9 citations in total.

Details

Primary Language Turkish
Journal Section Research Article
Authors

Mehmet Mendeş This is me

Publication Date January 1, 2004
Submission Date January 1, 2004
Published in Issue Year 2004 Volume: 10 Issue: 01

Cite

APA Mendeş, M. (2004). Ortanca Değerin Kullanılmasının ANOVA F Testi Bakımından Gerçekleşen 1. Tip Hata Olasılığı Üzerine Etkisi. Journal of Agricultural Sciences, 10(01), 20-23. https://doi.org/10.1501/Tarimbil_0000000863

Journal of Agricultural Sciences is published open access journal. All articles are published under the terms of the Creative Commons Attribution License (CC BY).