The process of research synthesis is examined in this paper. The results of a case study of what happens to the attitudes of six experts in the field of school desegregation and black achievement when they interactively review the relevant research literature are presented. The impact of these subjects' resulting reviews on the attitudes of a less expert audience (i.e. graduate students) is also examined. It is concluded that the attitude-relevant aspects of research reviewing are probably best understood by noting the special characteristics of the people and circumstances involved and by applying established principles to this unique situation. Attitudes are less likely to change if (as in the case of the research reviewers) there is a high degree of prior knowledge. In addition, the diversity of results often found in a set of empirical studies will inhibit attitude change in its consumers, further impeding consensus. And finally, even when there is agreement on the objective outcome of a set of studies, there may still be discrepancies resulting from subjective interpretations of what the measurements mean. Broader philosophic debates within the social sciences which are helping to define the limits of objectivity are considered and further studies based on the premise that knowledge synthesis is more meaningful if accompanied by descriptions of the process that brought the synthesis about are urged. (RDN)