Interpretative and Methodological Problems of Replicated Research.

The present paper discusses and illustrates some of the interpretative and methodological problems involved in replicated cross-national research by describing limited aspects of the set of replicated experiments conducted by the Organization for Comparative Social Research in Belgium, England, France, Germany, Holland, Norway and Sweden. Rather than repeating this material, we shall attempt to consider a single, key question of replicated research, if there are differences in the results of replicated experiments, how are such differences to be interpreted? Though systematic replications have been unfortunately rare in the social sciences, a sufficient number of such studies have accumulated to give us some experience with the interpretative problems involved in replication. When the results of replicated researches have been similar, the interpretation is usually clear and straightforward. Reproducibility has been demonstrated and, if the subject populations or the design of the experiment has been allowed to vary, there is evidence for the generalizability of the hypotheses involved. When the results of replicated researches have differed, however, interpretation has usually been exceedingly ambiguous.