The purpose of the present study was to apply the percentage of data points exceeding the median of baseline phase (PEM) approach for a meta-analysis of single-case experiments to compare the relative effectiveness of different kinds of reinforcers used in behavior modification. Altogether 153 studies were located, which produced 1091 effect sizes. The grand mean of the 153 PEM scores was .92. The main finding was that among the positive reinforcers, "activities" was the most effective (PEM score = .95) while "objects" and "edibles" were relatively less effective (PEM score = .85 and .83, respectively). The results of the present study are of importance in respect to the choice of reinforcer for use in intervention in behavior modification.