A study gathered data on the effectiveness of four phonological awareness measures in predicting reading outcome in first grade. The "Test of Non-Verbal Intelligence" (TONI) and the "Peabody Picture Vocabulary Test" (PPVT) were used to assess the cognitive abilities of 54 first-graders from a middle class elementary school in a midwestern city. Each subject was administered four tasks of phonological awareness during the first quarter of their first-grade year. Tasks were adaptations of tasks previously shown to be related to later reading ability. Outcome measures were reading tasks from the "Woodcock Reading Mastery Tests" (WRMT) and "Wide Range Achievement Test--Revised (WRAT-R)." These measures were administered during the fourth quarter of the first-grade year. Results indicated that two of the four phonological awareness tasks were more highly correlated to reading measures than were the TONI or PPVT-R. Results also showed that the "deletion" task accounted for a significant portion of word identification and word attack tasks. No other variable made a significant contribution for explaining the variance for the word identification and word attack. The "deletion" task also accounted for the most variance in WRAT-R task. Furthermore, the "deletion" task identified good readers with 85.7% accuracy and poor readers with 90.5% accuracy. These results support the effectiveness of specific phonological awareness measures in predicting first grade reading outcome. (Two tables of data are included; one appendix including deletion, categorization, blending task, and segmentation tasks is attached.) (PRA)