Minimization as an Alternative to Unrestricted Randomization in Educational Research

This paper describes an alternative to unrestricted randomization in experimental design, termed minimization, which can be used to achieve better balance of critical factors in small to medium-sized experimental studies. Several critical factors are controlled which are known to influence outcomes but which are not the foci of the study. Based on the critical factor values of previously entered participants, a new participant will be allocated into an experimental group to minimize imbalance across groups on critical factors. Minimization has been primarily employed in medical interventions and clinical trials. This article provides examples of the potential use of minimization in education, its advantages, and limitations. Minimization is one possible strategy to attain stronger validity in small to medium-sized experimental studies. (Contains 2 tables.)