AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING
Multiple linear regression techniques were employed to examine source selection in an Air Force procurement division. The experimental data was derived from an exercise which examined a part of the source selection process. Each exercise contained identical Request for Proposals. Subjects were asked to rate proposals with and without a policy prioritizing evaluation criteria. The results of the exercise were grouped three ways 1 by organization, 2 by supervisory level, and 3 by experience with source selection. All possible pairings within the above three categories were then compared to identify any significant differences between sub-category members. All groupings were investigated to determine whether or not each evaluation criteria contributed significantly to the rating a proposal received. Assimilation of the policy prioritizing evaluation criteria was examined. Last, the linear model was compared to a non-linear model to determine whether or not the linear model was adequate. The results suggest that source selection decisions are not similar across organizations within the Air Force division. Furthermore, subjects did not utilize all the information available to them in making decisions. People often chose to utilize only a part of the available information in arriving at a decision.