Whatever plan is designed, whether job-based or person-based, a crucial issue is the fairness of its administration. In research studies on job-based evaluation, it is treated as a measurement device; the research assesses its reliability, validity, costs, and its compliance with laws and regulations. In contrast, research on person-based structures tends to focus on their effects on behaviors and organization objectives and ignores questions of reliability and validity. A reliable evaluation would be one where different evaluators produce the same results. This is important because in practice, several different people usually evaluate jobs. The results should not depend on which person did the evaluation. Reliability can be improved by using evaluators who are familiar with the work and trained in the job evaluation process. Validity refers to the degree to which the evaluation assesses what it is supposed to—the relative worth of jobs to the organization. Many studies report that when different job evaluation plans are compared to each other, they generate very similar rankings of jobs but they disagree on how much to pay the jobs. For managing compensation the correct standard is the pay structure—what jobholders get paid—rather than simply the jobs’ rank order. Several methods are used to assess and improve employee acceptability. An obvious one is to include a formal appeals process. Employees who believe their jobs are evaluated incorrectly should be able to request reanalysis and/or skills reevaluation.