Proposed Nonparametric Test for the Mixed Two-Sample Design

A nonparametric test is proposed for a mixed design consisting of a paired sample portion and a two-independent-sample portion to test for a difference in treatment effects. The test is compared on the basis of estimated powers to a test developed by Dubnicka, Blair, and Hettmansperger (2002). Situations are found in which the proposed test has higher powers and situations are found in which the Dubnicka et al. test has higher powers.