supervision. Using the chi square test for differences in proportions it was found that in neither case was the numerical difference statistically significant. Examination of the two sets of whole job rankings will also yield information on the equality of jobs including supervisory duties. The rankings were computed by multiplying the stack number by 20 and adding the rank within the stack. Consequently, jobs with a lower rank are stronger jobs. Table 6 shows the mean whole job rankings assigned by classifiers and GAO managers for both samples. A similar pattern of results to the foregoing analyses can be seen. Females and minority persons have somewhat weaker jobs although, as before, the differences are not statistically significant. CONCLUSIONS 、 Since no significant patterns of differences were noted between white females and their male counterparts or nonwhites and white males, one may conclude that the Single Agency Series questionnaires do not support the hypothesis that nonwhites and women are not receiving similar quality work experiences as white males. However, there is a consistent pattern of somewhat lower job responsibilities among nonwhites and females. While this situation has not yet reached significant levels, it warrants further monitoring. The methodology used here has some potential as an unobtrusive way of detecting subtle differences in assigned job responsibilities. Because the data was collected for a different purpose, it was not subject to being shaded in a particular direction. Also, since the questionnaires were 35-533 - 79 46 (Pt. 2) reviewed and endorsed by the individual's supervisor the accuracy and validity of the data and results in enhanced. Despite these advantages, there are some drawbacks. Because the only source of information was the questionnaires, the results are not as conclusive as they might otherwise be. The classifiers indicated that some responses were not specific or detailed enough to allow them to make confident judgements. Future studies of this issue might be incorporated into an ongoing position classification survey program. This would retain the virtue of not concentrating attention on EEO aspects while gathering more conclusive information. |