Although many correctional education studies have identified various treatment programs as being effective for reducing recidivism, few, if any, of these studies appear to be above reproach when assessing their methodological vigor. This paper highlights the shortcomings in the current post-treatment quasi-experimental design primarily used to evaluate the effectiveness of various correctional education and vocational training programs. It is argued that this research design is inadequate to offer a true assessment of the impact of correctional education, especially if the outcome variable is recidivism, since numerous plausible variables that could impact program implementation, delivery, and retention, as well as post-release variables are missing from the model This survival analysis model, which evaluates only macro-level variables, often discounts alternative factors which might have caused a CE effect to have been dampened, thus in most of the studies, evidence that CE works to reduce recidivism is ambiguous. Further, replication of findings is virtually impossible using macro-level variables since their associated micro-level variables often are different for each location, facility, classroom, and instructor involved with the sites being studied. An argument is made to take a more holistic approach to evaluating the effects of correctional education programming, an approach which must start with the research funding agencies and a movement away from recidivism as the primary outcome variable.