Test-Taking Dispositions: A Missing Link?

Test-taking dispositions and motivation have been proposed as moderators of employment test validity. In a simulated, multiorganization employment system, scores on the Test Attitude Survey (TAS) were found to moderate the validities of both an ability and a personality test used to predict a perfor...

Full description

Saved in:
Bibliographic Details
Published in:Journal of applied psychology 1992-10, Vol.77 (5), p.629-637
Main Authors: Schmit, Mark J, Ryan, Ann Marie
Format: Article
Language:English
Subjects:
Ability tests
Aptitude tests
Attitude surveys
Attitudes
Biological and medical sciences
Employee testing
Employees
Employment
Employment Tests
Factors
Fundamental and applied biological sciences. Psychology
Human
Impacts
Job Performance
Motivation
Occupational psychology
Personality
Personality tests
Psychological tests
Psychology
Psychology. Psychoanalysis. Psychiatry
Psychology. Psychophysiology
Psychometrics
Statistical analysis
Statistical Validity
Studies
Surveys
Test Taking
Test Validity
Tests
Validity
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:Test-taking dispositions and motivation have been proposed as moderators of employment test validity. In a simulated, multiorganization employment system, scores on the Test Attitude Survey (TAS) were found to moderate the validities of both an ability and a personality test used to predict a performance criterion. The criterion-related validity of the personality test was found to be higher for a subsample with less positive test-taking motivation than for a subsample with higher test-taking motivation. An opposite effect was found among the subgroups for the ability test. Range restriction and the predictability of individuals with high and low test-taking dispositions were examined as explanations for the findings. Practice and negative feedback effects on test-taking motivation and dispositions were assessed. Implications for examining predictive versus concurrent validation designs are discussed.
ISSN:0021-9010
1939-1854
DOI:10.1037/0021-9010.77.5.629