Application of Principal Component Analysis in Determining Factor Weight in Point-Rating System of Job Evaluation

Abstract

The point-factor/point-rating system is the most popular method for job evaluation. While allocating the points corresponding to compensable factors through the consensus approach; judgmentally, it is typically a smooth process, provided the degree for each compensable factor is well defined. The allocation of relative weights to each compensable factor is a time-consuming and disputable aspect of the traditional point-rating system. In order to avoid the subjectivity of the weights allocated to factors, this paper aims to demonstrate the application of the principal component analysis (PCA) in determining the relative weights of the attributes/compensable factors in job evaluation. The PCA is also applied to arrive at the aggregated score of each job evaluated as per the point-rating system for determining the relative worth of jobs in an organization. The method has been tested by taking data of 15 jobs from a benchmark job evaluation point-rating study. The results show that the pattern of the scatter diagram with the scores of the jobs using both the systems is quite similar, and the PCA method discriminates the jobs slightly better than the traditional approach. The proposed method may also be called the hybrid method as it is a combination of both the traditional and statistical approach.