Journal of Kirkuk University For Administrative and Economic Sciences

The Use Of Some non-Parametric Methods In The Absence Of Homogeneity Of The Variation Of The Data For The Factorial Experiment (Application IN Veterinary Medicine)

Authors

Hadeel Imad Nasir
Kamal Alwan AL-Mashhadani

University of Kirkuk Journal For Administrative and Economic Science

Abstract
In this search we wanted to study the factorial experiments in order to test the significance of the effects of the main factors and the interaction between the levels of the factors at the same time , the test of the effects of significant of factors and interactions is by analysis of variance (ANOVA) requires that the experimental data meet the requirements of the variance analysis, but if the data does not meet the variance requirement, the heterogeneity requirement (one of the conditions or assumptions of the variance analysis) we obtained inaccurate and unreliable results, so it resorting to methods or other methods to reduce falling into this error and that give accurate results for this case are the methods nonparametric there are several nonparametric methods to analysis factorial experiment data: (Puri And Sen, Inverse Normal Transformation (INT) and Van Der Waerden) and applied to real data of the experiment obtained from the college of Veterinary Medicine University of Baghdad, where after testing data we found It suffers from the problem of heterogeneity .Then it found that the applying the non-parametric methods, the significant for some effects was obtained.

 

Keywords

Factorial Experiment
ANOVA Analysis
Parametric Test F
Nonparametric Tests
Journal of Kirkuk University For Administrative and Economic Sciences
Volume 10, Issue 1
Spring 2020
Pages 313-331

Home

Guide for Authors

Submit Manuscript

Reviewers

Contact Us