The purpose of this study was to determine the relative effectiveness of three different types of reinforcement strategies on spelling achievement among a sample of disadvantaged and non-disadvantaged fourth grade public school pupils. The three reinforcement strategies employed involved a concrete reward (candy), a token or symbolic reward (coupon with candy pay-off), and a social reward (reinforcing comments by the experimenter designed to approximate a typical student-teacher relationship). Treatment materials and spelling achievement tests were designed to approximate typical school subject matter. The hypothesis that among disadvantaged groups the concrete reward would have greater effect than either the token or social rewards was not confirmed. Factors relating to the school setting of the experiment or the basic characteristics of the experimental group may have influenced the results of the treatments. In addition, the results of the experiment may suggest that there are learning models other than the reinforcement type theory which are useful in the practical educational solution to the problem of low achievement among disadvantaged pupils. (Author/WR)