Abstract This article focuses on the validity of a prior study of the primary conclusion that "expectancy models were generally weak predictors of audit staff performance" as analyzed by Professor W. Bruce Johnson. He begins his comment by claiming that, while misquotation is an interesting way to develop one's position, it is clearly unacceptable in academic journals. He then appears to suggest that, since the expectancy models were found to perform poorly in the accounting environment and since these results compare favorably to those obtained in prior research efforts, one can only logically conclude that expectancy models are weak predictors of performance in all environmental settings. Johnson further argues that the methodology utilized a between subjects design, provides at best, a weak test of the theory's predictive ability. All in all, Johnson's comment does make a valid contribution, namely to clarify the secondary relationship between the expectance model and job satisfaction. However, his arguments are clearly insufficient to counter the primary study finding.
Kenneth R. Ferris (Sun,) studied this question.