ABSTRACT
This chapter examines inferential tests involving variances. Tests of variances are useful in two applications: as an inferential test by itself and as a test of the homogeneity of variance assumption for another procedure. The chapter explores both inferences about a single variance and inferences about two dependent variances. Dr. Abraham wants to determine if the variances of the achievement scores differ when children begin school in the fall as compared to when they end school in the spring. The chapter deals with concepts for testing inferences about variances, in particular, the sampling distributions underlying such tests. It explains several inferential tests of variances. Several tests have traditionally been used to test for the equality of independent variances. The chapter describes the somewhat cloudy situation that exists for the traditional tests. It provides details on two recommended tests, the Brown-Forsythe procedure and the O’Brien procedure.
