We evaluated the validity of DIBELS ("Dynamic Indicators of Basic Early Literacy Skills") ORF ("Oral Reading Fluency") for predicting performance on the "Florida Comprehensive Assessment Test" (FCAT-SSS) and "Stanford Achievement Test" (SAT-10) reading comprehension measures. The usefulness of previously established ORF risk-level cutoffs [Good, R.H., Simmons, D.C., and Kame'enui, E.J. (2001). "The importance and decision-making utility of a continuum of fluency-based indicators of foundational reading skills for third-grade high-stakes outcomes." "Scientific Studies of Reading," 5, 257-288.] for third grade students were evaluated on calibration (n[subscript S1] = 16,539) and cross-validation (n[subscript S2] = 16,908) samples representative of Florida's "Reading First" population. The strongest correlations were the third (February/March) administration of ORF with both FCAT-SSS and SAT-10 (r[subscript S] = 0.70-0.71), when the three tests were administered concurrently. Recalibrated ORF risk-level cut scores derived from ROC (receiver-operating characteristic) curve analyses produced more accurate identification of true positives than previously established benchmarks. The recalibrated risk-level cut scores predict performance on the FCAT-SSS equally well for students from different socio-economic, language, and race/ethnicity categories.