Hostname: page-component-745bb68f8f-b95js Total loading time: 0 Render date: 2025-01-08T12:05:26.158Z Has data issue: false hasContentIssue false
  • English
  • Français

The Problem of Classification of Personnel

Published online by Cambridge University Press:  01 January 2025

Robert L. Thorndike
Robert L. Thorndike*
Affiliation:
Teachers College, Columbia University
Get access Rights & Permissions [Opens in a new window]

Abstract

The personnel classification problem arises in its pure form when all job applicants must be used, being divided among a number of job categories. The use of tests for classification involves problems of two types: (1) problems concerning the design, choice, and weighting of tests into a battery, and (2) problems of establishing the optimum administrative procedure of using test results for assignment. A consideration of the first problem emphasizes the desirability of using simple, factorially pure tests which may be expected to have a wide range of validities for different job categories. In the use of test results for assignment, an initial problem is that of expressing predictions of success in different jobs in comparable score units. These units should take account of predictor validity and of job importance. Procedures are described for handling assignment either in terms of daily quotas or in terms of a stable predicted yield.

Type
Original Paper
Information
Psychometrika , Volume 15 , Issue 3 , September 1950 , pp. 215 - 235
Copyright
Copyright © 1950 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

*

Address of the President of the Division on Evaluation and Measurement of the American Psychological Association, delivered at Denver, Colorado, September 9, 1949.

References

* The term “success”, as it is used in this paper, may be interpreted quite broadly to include measures of job satisfaction as well as ratings of performance or measures of production.

* For the present, we are considering each of the two jobs to be equally important, and are dealing with the simple difference. It would also be possible to deal with a weighted difference, thereby attaching greater importance to one of the jobs than to the other.

* r1A, r2A, r2B are correlations of Tests 1 and 2 with job criteria A and B respectively.

† β1(A-B) and β2(A-B) are the regression weights for differential prediction.

‡ is the correlation between actual aptitude difference and predicted aptitude difference.

* This formula, which replaces an erroneous formula included in the original paper, was derived by William G. Mollenkopf. Its development is presented in Research Bulletin 50-9, Educational Testing Service, Princeton, N. J.

* Dr. P. J. Rulon has recently reported informally on the development of procedures for computing a multiple discrimlnant function which may eliminate the need for predicting success for separate job categories. The full report of this method will be awaited with interest.

* Hubert Brogden (An approach to the problem of differential prediction. Psychametrika, 1946, 11, 139-154) has approached the problem of a uniform score scale for predictions of success in different jobs from a somewhat different angle. He proposes translating everything into units of dollars saved by assigning Individual I, rather than an average individual, to Job A. However, he indicates that it will usually not be feasible to get a direct estimate of this dollar saving, and that one will have to rely upon subjective estimates of the type discussed in the present article. Brogden makes one point which is a desirable supplement to our discussion of a uniform scale for predictor scores. This is that one must take account of the extent to which efficiency varies from person to person. In some jobs the difference in level of performance may be relatively slight between the best individual and the average individual, while in other jobs it may be very great. The weighting factor for any job should take account not only of the importance of the job but also of the extent of individual differences in performance of it. The two judgments might be made separately, or they might be synthesized into a single compound judgment of the importance, to the success of the total organization, of the differences which are in fact found between individuals in the performance of that particular job.

* Personal communication to the author.

* Brogden, Hubert E. An approach to the problem of differential prediction. Psychometrika, 1946, 11, 139-154.