fplicate the existing grade structure. Remick ( 1978:97-105) findings that the new job scores resulting from other major comparable worth job evaluation studies also have high correlations with the original grades support these results. For example, in the 1976 Willis study of the State of Washington, 59 percent of the variation in the new job score could be explained by prior salary. In the Hay study of exempt staff of the University of Washington, the correlation between salaries and points for administrative and professional positions indicated that at least 72 percent of the variation in salaries could be explained by previous grade.
It is important to note that the overall effect of the new system was to replicate the old system and thereby perpetuate past wage discrepancies. However, even more important is the finding that this happened differently for men and women, as the benefit scores indicate: male-dominated jobs gained in the new ranking system, while female-dominated jobs lost ground. It has been demonstrated that the new FES had an adverse impact on those in female- dominated occupations, which could be used to establish a prima facie case under Title VII.
The statistical analysis in this chapter also provides a useful methodology for other researchers as they attempt to measure the effect of moving from one job evaluation plan to another.