Measurement Invariance in Performance Appraisal Ratings of US Army Special Forces Soldiers

measurement equivalence;performance appraisal;army special forces;performance ratings;CFA;DFIT;measurement invariance;IRT;item response theory;confirmatory factor analysis

Koommoo-Welch, Penny ; Dr. Michael G. Sanders, Committee Member,Dr. Mark A. Wilson, Committee Chair,Dr. S. Bart Craig, Committee Member,Dr. Donald W. Drewes, Committee Member,Koommoo-Welch, Penny ; Dr. Michael G. Sanders ; Committee Member ; Dr. Mark A. Wilson ; Committee Chair ; Dr. S. Bart Craig ; Committee Member ; Dr. Donald W. Drewes ; Committee Member

University:North Carolina State University

关键词: measurement equivalence;    performance appraisal;    army special forces;    performance ratings;    CFA;    DFIT;    measurement invariance;    IRT;    item response theory;    confirmatory factor analysis;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/1666/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

The purpose of the present study was to examine the equivalence of mental models of job performance between rater groups among US Army Special Forces Soldiers. Performance appraisals are often completed in organizations by individual raters, whose ratings are then compared to one another in order to make inferences of the ratee's performance on the job. Disagreements in ratings can lead to erroneous conclusions unless it is first established that comparisons between the rater groups are appropriate. Ratings of soldiers by two supervisory rater groups (N = 1052 and N = 910) on an appraisal instrument designed specifically for the Special Forces were examined. An exploratory factor analysis of the ratings indicated a four-factor model, which was then subsequently used to test for measurement invariance between the rater groups using multiple-group confirmatory factor analysis and item response theory. Fit indices indicated reasonable fit of the model, and ratings were concluded to be invariant at the rater group level of analysis, indicating that the rater groups refer to similar mental models of performance when rating individual soldiers.