Testing differences between two samples variances

This can be tested with a variance ratio test:

$$F = \frac{s^{2}_{2}}{s^{2}_{1}} = \frac{s^{2}_{1}}{s^{2}_{2}} $$

The largest value is placed in the numerator. This test is compared to the F distribution for a significance decision.

[Image]
Figure 1. Example of an F distribution illustrating the critical value for an a = 0.05, v1 = 10, v2 = 10.

To evaluate the results of a test, if the value from the formula above is between 0 and the critical value ( the white area under the curve) the two variance values are not significantly different at the specified α level. If the value from the formula is greater than the critical value (the red area under the curve) the two variance values are significantly different at the specified α level.

For table of F critical values at the α= 0.05 level.


Also See:

Chapter 9 - Significance of a Difference between Two Means pages 108-124 in:

Phillips, J. L. 2000. How to think about statistics. W. H. Freeman and Co. New York. 202 pp. ISBN 0-7167-3654-3

Chapter 9 - Two-Sample Hypotheses pages 126-130 in:

Zar, J. H. 2007. Biostatistical Analysis. Prentice-Hall, Inc. Englewood Cliffs, New Jersey. 718 pp.


Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 3.0 United States License.

Author: Dr. David R. Larsen
Created: July 19, 2000
Last Updated: December 14, 2019