Educators have always been interested in identifying and implementing effective programs. In today’s educational environment, that interest is amplified by the fact that schools and school districts are increasingly expected to select programs with a strong research base (Lynch and O’Donnell, 2005; Mowbray, et al., 2003). In this context, the What Works Clearinghouse and a recent National Research Council panel, have called for more rigorous studies (WWC, 2004a, 2004b; NRC, 2004) to identify effective instructional materials and other programs. Researchers have responded, but regrettably, even as they use more rigorous designs to study interventions and describe and measure their impacts on student outcomes, they overlook the importance of measuring and describing the implementation of the treatment itself (Hall & Loucks, 1977; WWC, 2004a, 2004b; NRC, 2004). Without specific, clear descriptions and measurement of implementation, it is impossible to know whether ill-achieved outcomes are due to an inadequate model of change in a program, or due to poor or incomplete implementation (Fullan, 1983; Lynch & O’Donnell, 2005; Wang et al., 1984; Ruiz-Primo, 2005).Penuel and Means (2004) assert that researchers can use implementation data to “identify competing hypotheses” as well as flawed assumptions in the program design (p. 332).