encompassed the entire nine-point appraisal scale. The distribution in ratings
between the two time periods was signiWcant.
Still another training approach, consistent with the cognitive processes required
in a performance appraisal, is frame-of-reference training or FOR (Bernardin et al.
2000 ). The purpose of FOR training is to calibrate raters so that they agree on (a)
how to match speciWc ratee behaviors to the appropriate performance dimensions,
(b) the eVectiveness levels of alternative behaviors, and (c) the rules for combining
individual judgements into a summary evaluation for each performance dimension
(Sulsky and Day 1992 ). To the extent that FOR training helps raters formulate correct
impressions of a ratee’s performance on each performance dimension, rating ac-
curacy increases. This occurs even if raters forget speciWc performance information
and rely upon their overall impressions of a ratee’s performance, as is predicted by
cognitive processing models (Sulsky and Day 1994 ; Noonan and Sulsky 2001 ).
Implicit person theory (IPT), that is, beliefs regarding the malleability of behav-
ior, aVects appraisals of others (Heslin et al. 2005 ). Entity theorists believe that an
individual’s personal attributes are largelyWxed. This leads them to quickly form
strong impressions of an individual that they resist reversing, despite subsequent
contradictory information. The IPT of incrementalists, on the other hand, is that
personal and situational determinants of an employee’s behavior are dynamic.
Thus they reconsider their initial impressions if new information warrants it.
Failure to recognize a signiWcant decrease in the performance of a surgeon or
a pilot, for example, could be catastrophic. Similarly, a failure to acknowledge a
signiWcant improvement in behavior can lead to employee resentment and with-
drawal. Heslin et al. showed how Pratkanis and Aronson’s ( 2001 ) self-persuasion
methodology can be used to train entity theorists to adopt an incremental IPT that
is sustained over time.
Martell et al. ( 1995 ) found that memory of job-related behavior is aVected by
performance expectations. Raters who believe that a person has performed well vs.
poorly report observing more eVective and less ineVective behaviors regardless of
whether the behaviors actually occurred. Source monitoring training corrects this
error. Martell and Evans ( 2005 ) trained raters to report only behaviors that evoke
detailed memories and to suppress reporting behaviors that are based on feelings of
familiarity when completing a behavioral checklist. Consequently, raters were
capable of identifying how they remember what they remember in doing
a performance appraisal. Distinguishing between the two sources of memory
judgements reduced the biasing eVect of rater expectations of an employee.
- 1 Context
Often overlooked in training programs is the social and political context
of appraisal. Training focuses primarily on ability rather than the motivation of
performance management 373