The purpose of this article is to present a model for screening for twice-exceptional status (i.e., gifted students who have a learning disability). Curriculum-based measures (Monitoring Instructional Responsiveness: Reading and Monitoring Instructional Responsiveness: Math) were administered to 1,242 third-grade students within a Response to Intervention paradigm. When gifted status is tentatively defined as high performance (i.e., 84th percentile and higher) on a Monitoring Instructional Responsiveness reading probe, 5.48% of students exhibited deficits in (math) performance consistent with a significant discrepancy between reading and math (i.e., reading score -- math score); 4.83% exhibited a discrepancy in reading (i.e., math score -- reading). These values are based on "observed" scores using the following formula to define a discrepancy: 1.5(SD) × SEe. Only 2.1% exhibited a math discrepancy and 1.13% a reading discrepancy based on "predicted" scores, which takes regression to the mean into account. Using various cut score criteria, practitioners can select from less than 1% to about 10% for screening purposes. When using predicted (rather than observed) scores and more stringent cut score criteria, percentages decline, as expected. Recommendations for using this process for screening are provided, as are implications for best practice, particularly the impact of using more or less conservative criteria for screening twice exceptional students.