Testing the Generalizability of Indirect Range Restriction Corrections

Range restriction is a common problem in personnel selection and other contexts in applied psychology. For many years researchers have used corrections that assume range restriction was direct, even when it was known that range restriction was indirect. Hunter, Schmidt, and Le (2006) proposed a new...

Full description

Saved in:
Bibliographic Details
Published in:Journal of applied psychology 2014-07, Vol.99 (4), p.587-598
Main Authors: Beatty, Adam S, Barratt, Clare L, Berry, Christopher M, Sackett, Paul R
Format: Article
Language:English
Subjects:
Adult
Alternatives
Applied Psychology
Biological and medical sciences
Comparative analysis
Error of Measurement
Fundamental and applied biological sciences. Psychology
Generalizability
Humans
Information
Methodology
Models, Statistical
Occupational psychology
Orientation. Selection. Evaluation
Personnel
Personnel management
Personnel Selection
Personnel Selection - statistics & numerical data
Psychology. Psychoanalysis. Psychiatry
Psychology. Psychophysiology
Restriction
Structural models
Studies
Testing
Usefulness
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Range restriction is a common problem in personnel selection and other contexts in applied psychology. For many years researchers have used corrections that assume range restriction was direct, even when it was known that range restriction was indirect. Hunter, Schmidt, and Le (2006) proposed a new correction for cases of indirect range restriction that greatly increases its potential usefulness due to its reduced information requirements compared to alternatives. The current study examines the applicability of Hunter et al.'s correction to settings where its assumed structural model is violated by including the measures that are to be involved in corrections in the original selection composite. We conclude that Hunter et al.'s correction should generally be preferred when compared to its common alternative, Thorndike's Case II correction for direct range restriction. However, this is due to the likely violation of one of the other assumptions of the Hunter et al. correction in most applied settings. Correction mechanisms and practical implications are discussed.
ISSN:0021-9010
1939-1854
DOI:10.1037/a0036361