Electronic Theses and Dissertations

Identifier

1242

Date

2014

Document Type

Dissertation (Access Restricted)

Degree Name

Doctor of Philosophy

Major

Psychology

Concentration

Experimental Psychology

Committee Chair

William Dwyer

Committee Member

Randy Brou

Committee Member

William Farmer

Committee Member

Randy Floyd

Abstract

The purpose of the study was twofold: (1) determine the degree to which the “empirical” weights of the FMS variables were consistent with the “current” weights attributed to each component of the FMS, when they were examined at the community and rating levels, and (2) If differences between the “empirical” weights and “current” weights were found at the community level, then determine if differential scores in the promotion outcome were due to FMS components, AFQT, years in service, and/or the effects of moderator variables race and gender. The Nuclear community (critical ratings) and Administration community (non-critical ratings) were specifically chosen to examine if the current FMS promotion system functions effectively across two different types of communities. Support was found for primary hypotheses (hypothesis 1 and 2). Hypothesis 1 stated that the “empirical” weights of the FMS variables were not consistent with the “notional” weights, currently used by the NEAS, and varied by community, rating, and pay grade level. The results from the analyses confirmed this hypothesis and showed that the actual contribution of the FMS components (SS, PMA, SIPG, PNA, ED, AWD) are not reflective in the current component weighting algorithm. The analyses from the regression models provided a statistically optimal combination whereby the SS and PMA accounted for the majority of the variance found in the FMS composite (supported Hypothesis 2). The secondary analyses supported that the full model containing all the predictors were statistically significant, indicating that the model was able to distinguish between participants who were promoted and those who were not promoted. The results from the secondary analyses indicated subgroup differences in scores related with gender and race; however, the Cohen’s effect size statistic showed that the magnitude of subgroup differences varied by community and pay grade. Despite weighting differences found across communities and rating levels, incorporating community-specific FMS algorithms was not warranted at the E4 to E7 pay grades. The current weighted enlisted advancement system works effectively at the generalized pay grade level. Because the Navy uses a multiple predictor model, using cognitive tests in combination with other non-cognitive predictors, it mitigates the problems associated with subgroup disparities without compromising the overall validity of the promotion system.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Download

Share

COinS