“A Causal Analysis of Cognitive Ability, Job Knowledge, Job Performance, and Supervisor Ratings”, 1983 (; similar):
Through a meta-analysis of 14 studies, Dr. Hunter investigates the relationships among three variables: ability, job knowledge and performance. Performance is measured in two ways, work sample tests and supervisory ratings. Two causal models are presented which depict the possible relationships among the three variables. The first model suggests a direct impact of ability on performance, on job knowledge, and on supervisor ratings. This implies that there is an indirect relationship impact of job knowledge on supervisory ratings. The alternative model suggests that job knowledge is directly related to supervisory ratings. The results of the path analysis provide support for the latter model.
Dr. Guion states that Hunter’s findings provide evidence for the validity of ratings. However, the model is incomplete. He suggests that the model should be enlarged to include such exogenous variables as rater characteristics, ratee characteristics, and context factors.