Premium
Variables affecting leniency, halo and validity of self‐appraisal
Author(s) -
Fox Shaul,
Caspy Tamir,
Reisler Avner
Publication year - 1994
Publication title -
journal of occupational and organizational psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.257
H-Index - 114
eISSN - 2044-8325
pISSN - 0963-1798
DOI - 10.1111/j.2044-8325.1994.tb00548.x
Subject(s) - psychology , halo effect , convergent validity , rating scale , halo , social psychology , external validity , sample (material) , promotion (chess) , inclusion (mineral) , test validity , performance appraisal , incremental validity , response bias , psychometrics , clinical psychology , developmental psychology , chemistry , physics , management , chromatography , quantum mechanics , galaxy , politics , political science , economics , law , internal consistency
Three experimental manipulations were examined for altering leniency, halo and concurrent validity of self‐ratings: (a) cautionary instructions regarding cross‐checking of self‐evaluations, (b) inclusion of dimensions less relevant to the immediate assessment setting, and (c) the use of positively toned, unbalanced scales. A sample of 275 Israeli police officers rated themselves at the end of an assessment centre designed to assess supervisory potential. Rating forms contained 12 dimensions of high and low relevance to promotion potential. Half of the subjects were prewarned that their ratings would be matched with other data. In an overlapping manipulation, half rated themselves on conventional balanced scales while the other half were given positively toned, unbalanced scales. It was found that leniency and halo were reduced by unbalanced scales and the introduction of less relevant dimensions, but not affected by cautionary instructions. These decreases in bias were accompanied by increased convergent validity ( vis‐à‐vis peer and supervisory ratings) in the case of unbalanced scales. Discussion focuses on the role of self‐enhancement motives in leniency effects, as well as the relation between bias and validity. Practical suggestions for improving self‐rating procedures are offered.