Nonconcurrent Multiple Baseline Designs and the Evaluation of Educational Systems

Abstract

Important findings are often a balance between the rigor of the experimental design and innovativeness of the experimental question. One broad topic area that has received a great deal of discussion, but little empirical study, is the evaluation of educational systems. Experimental designs that permit the analysis of practices used by state education agencies, local education agencies, and schools have the potential for yielding socially significant findings that could improve education. In this article we discuss the use of nonconcurrent multiple baseline designs as an option for studying the activities and effects of educational programs. Nonconcurrent multiple baseline designs stagger the timing of baseline-to-intervention changes across various entities, but the baselines and intervention phases are not contemporaneous across each of the tiers. Although considered less rigorous than concurrent multiple baseline designs, nonconcurrent designs have a degree of flexibility that may allow for their use in studying complex social contexts, such as educational settings, that might otherwise go unanalyzed.