Analytic Methods for Adjusting Subjective Rating Schemes

by Richard V.L. Cooper, Gary R. Nelson

Full Document

FormatFile SizeNotes
PDF file 1.9 MB

Use Adobe Acrobat Reader version 10 or higher for the best experience.

Subjective evaluations of individual performances by supervisors are subject to bias. It is important to correct for biases in order to more accurately measure the effects of specific variables on individual performance. This report develops statistical and econometric techniques for correcting biases in models of individual performance using a variant of the classical linear regression model. A multiscale model is proposed to deal with two types of bias: location bias when an individual's performance is systematically overestimated or underestimated, and scale bias when differences among individuals rated are exaggerated or minimized. Several specific multiscale estimating techniques are developed, including equal total variance, equal residual variance, maximum likelihood, and least squares. Finally, the multiscale estimators are applied to the problem of estimating the cost of on-the-job training in the military. The multiscale model can be applied to a wide variety of estimating problems where observations can naturally be categorized into specific subgroups.

This report is part of the RAND Corporation Report series. The report was a product of the RAND Corporation from 1948 to 1993 that represented the principal publication documenting and transmitting RAND's major research findings and final research.

Permission is given to duplicate this electronic document for personal use only, as long as it is unaltered and complete. Copies may not be duplicated for commercial purposes. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. For information on reprint and linking permissions, please visit the RAND Permissions page.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.