Abstract
The Rotter Incomplete Sentences Blank (RISB; Rotter, Lah, & Rafferty, 1992; Rotter & Rafferty, 1950) was administered to African American (n = 94) and White (n = 100) members of fraternal organizations at a large Midwestern university to explore possible influences of racial group and socially desirable response tendencies. The RISB protocols were reliably scored (interrater agreement = 92%) by 3 independent scorers blind to participant racial membership. The mean scores of the 2 racial groups were not significantly different (p > .05). However, using the traditional categorical placement scores resulted in a disproportionate assignment rate for African American scores to the maladjusted group than would be anticipated by mere chance.