The urban, ethnically Hawaiian child typically experiences great difficulty in learning to read English. In order to determine whether phonological confusion is a source of dialectical interference, the Kamehameha Early Education Program (KEEP) Phone Discrimination Test (KPDT) was developed for the one hundred twelve students in the KEEP school (K-3). This research report describes the development of the procedure for item inclusion and for administration of the test. The research sought to answer the following questions: (1) whether bidialectal confusion difficulties should be assumed, (2) whether all confusions of this sort are idiosyncratic in nature, and (3) whether there is a need for a program to eliminate confusions due to differing phoneme inventories of an idiolectal or dialectal nature. Eight Hawaiian-Creole-speaking children selected by KEEP personnel were the subjects in the development of the test. The final version concentrates on five phoneme pairs which appear to be sources of allophonemic confusion for the children tested. It appears to test (at the allophonemic level) points of overall phonemic confusion which could be predicted to be sources of difficulty in learning phoneme-grapheme correspondence. This report includes a test sample and directions for its administration and use. (AMH)