Call Me Old Fashioned - Is My Job Analysis Accurate or Not?

Abstract

As a process designed to collect information about jobs, job analysis is one of the most fundamental aspects of personnel psychology. It forms the foundation upon which almost every other human resource management component is built, including selection, compensation, performance appraisal, and training program development. Despite the considerable evidence of human fallibility in other judgment processes, many have followed the implicit assumption that job analysis information is accurate without actually examining this proposition. This study considers two potential sources of job analysis rating inaccuracy — the source of the ratings and the type of instrument utilized to collect ratings. By utilizing less job-familiar job analysis raters and shorter, more holistic job analysis instruments, industrial-organizational psychologists have attempted to attenuate the time and costs associated with the job analysis process; however, findings regarding the reliability and accuracy of such practices are questionable. Hypotheses tested in the current study indicated that decomposed measures of job behavior converged to a greater degree with an external job analysis than did holistic measures. Interrater agreements for all types of raters and across all types of instruments are concluded to be inadequate. Potential explanations from the cognitive and social psychological domains for these findings are conjectured and directions for future research are noted.