Computer simulations in evaluation research are useful for (1) improving student understanding of basic research principles and analytic techniques; (2) investigating the effects of problems that arise in the implementation of research; and (3) exploring the accuracy and utility of novel analytic techniques applied to problematic data structures. This article describes these uses of microcomputer simulations for the context of human service program evaluation. Simple mathematical models are described for the three most commonly used human service outcome evaluation designs: the pretest-posttest random ized experiment, the pretest posttest nonequivalent groups design; and the regression- discontinuity design. The models are translated into a single microcomputer program that can be used to conduct the simulations. Examples of the use of this program on an IBM PC microcomputer are provided to illustrate the three uses of the simulations described. The article concludes by arguing that simulations need to utilize experimental design principles when rigorous, definitive results are desired, but that simulations may have great potential value as an exploratory or teaching tool in human service research.
