Peer evaluations have historically shown high predictive validity, but the
reason for this strength has not been clear. This research used an assessme
nt center and subsequent training program to investigate the hypothesis tha
t interpersonal performance and motivation are two key dimensions of perfor
mance that may account for the historical strength of peer assessments. Con
sistent with previous research, results showed that peer rankings from the
assessment center predicted final training outcomes better than did staff r
atings. Congruent with our hypothesis, results from the training program de
monstrated that, when rating a student's expected future on-the-job perform
ance, peers placed significantly more importance on interpersonal performan
ce and motivation than did staff, and they placed significantly less import
ance on task performance. Additional longitudinal research is needed to exp
licitly link peer evaluations of interpersonal skills and performance motiv
ation with future job performance.