Testing multiple dispersion effects in unreplicated fractional factorial designs

in unreplicated 2(k-p) designs, the assumption of constant variance is comm only made. When the variance of the response differs between the two levels of a column in the effect matrix, that column produces a dispersion effect . In this article we show that two active dispersion effects may create a s purious dispersion effect in their interaction column. Most existing method s for dispersion-effect testing in unreplicated fractional factorial design s are subject to these spurious effects. We propose a method of dispersion- effect testing based on geometric means of residual sample variances. We sh ow through examples from the literature and simulations that the proposed t est has many desirable properties that are lacking in other tests.