The main goal of this report is to present evidence on the predictive validity of an English language arts (ELA) performance assessment (PA) administered in Grades 2-9 in a large urban school district. To account for the hierarchical structure of the data (students are nested within schools), we employed hierarchical linear modeling (HLM) to distinguish individual and aggregated explanatory variables. Based on a sub-sample of 5,427 students, we found that students' 2001 ELA PA scores were predictive of their probability of passing the California High School Exit Exam (CAHSEE). We also found a significant correlation between student performances on the ELA performance assessment and other standardized tests. We believe that the ELA PA may be a dependable and useful indicator to identify at-risk students. (Contain 3 figures and 7 tables.) [This report was supported in part by the Office of Educational Research & Improvement, U.S. Department of Education.]