Response to intervention (RTI) has emerged as a promising, but possibly vulnerable, alternative to the current model of identification and eligibility assessment in special education. Nested within a simple and compelling structure for RTI are many questions and challenges that make its future as "policy" at-risk unless these issues can be satisfactorily resolved. As it stands, RTI is likely to improve many decisions about educational interventions, especially when combined with universal screening procedures. In a full RTI implementation, potential benefits may include (a) increased achievement schoolwide, because struggling children are identified proactively and immediate help is provided to determine whether or not more intensive help is needed, as well as (b) a coherent and flexible system of services. Whereas many of the components of RTI are well established, RTI needs further operationalization, development, standardization, and field-testing. All first-and second-grade children (N = 182 children; 364 potential reading or math cases) at an elementary school were screened using four measures (curriculum-based measurement, Brigance subtests, state reading test, and teacher identification). Decision rules were applied to the screening data to select those children who were at-risk academically. Each of these children (n= 101 reading and math cases) participated in an individual curriculum-based assessment (CBA) in reading or math. Of these cases, 32 met criteria to participate in individual intervention (16 math and 16 reading interventions). Standard protocol intervention was implemented for 5 to 9 intervention sessions or until children met intervention success criteria. All second-grade children were administered the Iowa Test of Basic Skills. Data were generated to examine various criteria for judging intervention responsiveness and to raise empirical questions about the reliability and accuracy with which decisions could be made under an RTI model. (Contains 5 tables.)