A Comparison of Power of the Test of Homogeneity of Variances

Tidarat Phanomsophon, Apinya Hirunwong, Boonorm Chomtee


The objective of this research is to compare power of the test of homogeneity of variances of 3, 4 and 5 populations by Levene’s test (W) Brown-Forsythe’s test (BF) and Bhandary and Dai’s test (BD). The methods are compared by considering the ability to control the probability of type I error and power of the test. The studied use simulated normal distribution data in small, medium and large size . The results show that Bhandary and Dai’s test can control probability of type I error for all cases and It has highest power of the test.


Keywords :  homogeneity of variances, Levene’s test, Brown-Forsythe’s test, Bhandary and Dai’s test

Full Text:



Bhandary, M., & Dai, H. (2009). An alternative test for the equality of variances for several populations

when the underlying distributions are normal. Communications in Statistics – Simulation and

Computation, 38(1), 109-117.

Brown, M.B., & Forsythe, A.B. (1974). Rubust tests for the equality of variances. Journal of the American

Statistical Association, 69, 364-367.

Cochran, W.G., (1954). Some methods for strengthening the common tests. Biometrics, 10, 417-451.

Games, P.A., Winkler, H.B., & Probert, D. A. (1972). Robust tests for homogeneity of variance. Educational

and Psychological Measurement, 32, 887-909.

Levene, H. (1960). Robust tests for equality of variances. Contributions to probability and statistics: essay in

honor of Harold Hotellings, 278-292.

Thukaeow, P. (2009). Comparison on Homogeneity Test of Variances. Retrieved February 2, 2017, from


Smalheiser, N.R. (2017). Chapter 11: ANOVA. Data Literacy, 149-155.


  • There are currently no refbacks.