An internal bias analysis of a job evaluation instrument

Abstract

[Correction Notice: An erratum for this article was reported in Vol 70(3) of Journal of Applied Psychology (see record 2009-19520-001 ). In Table 2, for the female sex-typed jobs, there were errors in the correlations between Scale 6, Physical Effort, and Scales 11-15, and in the correlations with Total Points. The correct correlations are provided in the erratum. Only the change for the correlation between Physical Effort and Total Points changes the results: The female sex-typed correlation is now significantly higher than the male sex-typed correlation.] PhD-level industrial and organizational graduate students (2 males and 2 females) rated 105 male sex-typed and 105 female sex-typed job descriptions on a 15-scale point job evaluation instrument (The Comprehensive Job Evaluation Technique) as a means of detecting sex bias in the internal structure and choice of scales. Analysis, including reliability, factor analysis, partial correlation, and scale-total correlations, showed that the instruments as a whole differentiated male and female sex-typed jobs equally well. It is suggested that internal bias analysis can provide a useful technique for exploring internal sex bias in the development and selection of job evaluation instruments. It is concluded that the principles of internal bias analysis are the same as the principles for constructing a good test. An example of an education scale and definitions of all comprehensive job evaluation technique scales are appended. (47 ref) (PsycInfo Database Record (c) 2025 APA, all rights reserved)