The paper reviews the debate on dissensus versus consensus in occupational prestige scales. Varieties of dissensus are distinguished, and it is argued that dissensus related to the social characteristics of raters is the important form. Previous evidence of dissensus thus understood has tended to involve data bases restricted to small numbers of raters or occupation titles. The present study attempts to fill this gap by a reanalysis of national prestige surveys collected bynorc.The most striking finding is that dissensus within social groups is particularly high among lowsesstrata and among blacks. The role of random error in this pattern is discussed.