You are here

UCF Digital Collections ยป ETD Collection

ISSUES IN VALIDITY GENERALIZATION: THE CRITERION PROBLEM

Download pdf | Full Screen View

Date Issued:
2010
Abstract/Description:
Schmidt and HunterÂÂ's validity generalization model poses seven sources of error variance affecting validation studies. Of the seven sources of error variance, only four sources have been tested. This study looks at an additional source of error variance, the difference between studies in the amount and kind of criterion contamination and deficiency, as proposed by Schmidt and Hunter. The current study proposes a method of evaluating criterion contamination and deficiency in criterion measures in order to minimize their effects on the relationship between criterion and predictor measures. Two unique criteria are used including a traditional subjective measure of current performance and a non-traditional subjective measure of expandability (future performance). Data from 378 employees from a large international financial institution were used to test the proposed method. Results do not support the hypotheses. Single criteria predicted the same or better than the combined criteria, suggesting that the criterion problem was not addressed. Possible reasons for these findings are discussed. An unexpected finding supports the utility of personality measures compared to cognitive ability measures. The study concludes with a discussion of the implications and limitations of the study as well as directions for future research.
Title: ISSUES IN VALIDITY GENERALIZATION: THE CRITERION PROBLEM.
29 views
15 downloads
Name(s): Hodge, Raquel, Author
Wooten, William, Committee Chair
University of Central Florida, Degree Grantor
Type of Resource: text
Date Issued: 2010
Publisher: University of Central Florida
Language(s): English
Abstract/Description: Schmidt and HunterÂÂ's validity generalization model poses seven sources of error variance affecting validation studies. Of the seven sources of error variance, only four sources have been tested. This study looks at an additional source of error variance, the difference between studies in the amount and kind of criterion contamination and deficiency, as proposed by Schmidt and Hunter. The current study proposes a method of evaluating criterion contamination and deficiency in criterion measures in order to minimize their effects on the relationship between criterion and predictor measures. Two unique criteria are used including a traditional subjective measure of current performance and a non-traditional subjective measure of expandability (future performance). Data from 378 employees from a large international financial institution were used to test the proposed method. Results do not support the hypotheses. Single criteria predicted the same or better than the combined criteria, suggesting that the criterion problem was not addressed. Possible reasons for these findings are discussed. An unexpected finding supports the utility of personality measures compared to cognitive ability measures. The study concludes with a discussion of the implications and limitations of the study as well as directions for future research.
Identifier: CFE0003158 (IID), ucf:48609 (fedora)
Note(s): 2010-05-01
M.S.
Sciences, Department of Psychology
Masters
This record was generated from author submitted information.
Subject(s): criterion problem
contamination
deficiency
situational specificity hypothesis
multiple criteria
Persistent Link to This Record: http://purl.flvc.org/ucf/fd/CFE0003158
Restrictions on Access: campus 2011-10-27
Host Institution: UCF

In Collections