Journal of Applied Sciences1812-56541812-5662Asian Network for Scientific Information10.3923/jas.2009.2916.2924KaramanF.122009916In this study, Klotz’s test for heterogeneity of variance is generalized to factorial designs. Although, Levene’s test, the jacknife and Fligner and Killeen’s test for heterogeneity of variance were generalized to factorial designs, Klotz’s test was not studied. The performance of Klotz’s test is compared with other previously studied tests. An application and comparison of these analyses to the 2 by 2 factorial design are examined in detail. A simulation study was performed for 10000 data sets to compare power and robustness properties of the tests. It is observed that as sample size increased and as the difference in variances increased all tests have higher power. According to these simulations, Klotz’s test can particularly be recommended for symmetric parent distributions. For skewed parent distribution like chi-square, Klotz’s test should not be used since it is not robust. Data from a multifactor agricultural system were used as an example to illustrate the usefulness of these tests.]]>Conover, W.J.,1980Conover, W.J., M.E. Johnson and M.M. Johnson,1981Conover, W.J. and R.L. Iman,1981Fligner, M.A. and T.J. Killeen,1976Klotz, J.,1962Miller, R.G.,1968Milliken, G.A. and D.E. Johnson,1984Ozaydin, F., D.M. VanLeeuwen, C.S. Miller and J. Schroeder,1999Anderson, M.J., 2006McGrath, R. and D. Lin, 2002