AbstractTwo studies were designed to test hypotheses derived from Locke's (1969, 1976, 1984) model of job satisfaction. Consistent with the model, both studies found that perceived have‐want discrepancy scores were powerful predictors of facet job satisfaction. In addition, facet importance tended to moderate this relationship in the manner predicted by Locke. Workers who viewed a job facet as having high importance were more satisfied with a small perceived have‐want discrepancy and more dissatisfied with a large discrepancy than workers who viewed the facet as having low importance. Finally, as expected, facet importancefailedto moderate the relationship between facet satisfaction and overall job satisfaction. This finding supports Locke's proposition that facet satisfaction scores are ‘implicitly weighted’ by facet importance. Implications and directions for future research are di