The lack of correspondence between job satisfaction attitudes and behavioral indicators of job satisfaction exemplifies the more widely recognized problem of attitude-behavior inconsistency. While various theoretical and methodological explanations for this inconsistency have been offered, it is posited that this inconsistency can be reduced through a more accurate (valid) assessment of the cognitive component of attitudes. The notion of a saliency criterion in the evaluation of the cognitive component of job satisfaction is suggested as such a technique for achieving more accurate assessment. The concept of saliency itself as well as the notion of a saliency criterion is explored using a sample of 72 blue collar women. The concept is found to be both viable and relatively stable for various subgroups within the sample, as defined by several factors external to the job. Job satisfaction scales using specified saliency criteria are compared with scales which do not consider saliency on the basis of their ability to correlate with a behavioral measure of job satisfaction (absenteeism). The findings, with certain delineated reservations, indicate that scales which employ saliency criteria are better correlates of behavior. Saliency thus may be viewed as at least a partial solution to the problem of attitude-behavior inconsistency. / Master of Science