The linear programming alternative to policy capturing for eliciting criteria weights in the performance appraisal process

Abstract

An important aspect of management is the periodic performance appraisal (PA) of subordinates. This paper focuses on inferring the criteria employed and weights attached to them by an assessor in any PA process. Linear programming (LP) is proposed as an alternative to policy capturing (PC) as the inference mechanism. The LP approach is illustrated and contrasted with regression-based PC approaches. We show that, at a minimum, LP provides facilitating inputs to complement PC.