When it comes to performance appraisals, notions of fairness are all over the map, study finds
February 22, 2016
For more information, contact: Ben Haimowitz , (718) 398-7642, firstname.lastname@example.org
Little consensus about what makes for justice and what doesn't
Annual performance appraisals, those dreaded rites of
organizational life, have been getting the brush-off lately, with such
behemoths as IBM, Microsoft, and General Electric abandoning the practice.
What's wrong with these rituals? While not reaching any overall verdict about
their value, new research in organizational justice finds that notions of what makes
or doesn’t make performance appraisals fair are all over the map.
The research in the forthcoming issue of the journal Academyof Management Discoveries starts by noting that "the
perceived fairness of performance appraisals has been shown to influence
employees' commitment to their organization, their subsequent performance, and
even their mental states such as anxiety and depression." It then proceeds
to document "the idiosyncratic nature of forming global fairness
judgments" and to reveal sharp divergence among 49 seasoned administrators
over "how important specific aspects of justice are to them."
Comments Hayley German of the London School of Economics and
Political Science, who carried out the study with Marion Fortin of the
University of Toulouse and Daniel Read of Warwick Business School, "The
fact that experienced administrators differ sharply in how they evaluate the
fairness of the same appraisal suggests why this can be a potential minefield
for employers. On the basis of our findings, it comes as no great surprise that
annual performance appraisals have been losing favor."
To illustrate the wide disparity in approaches to this exercise,
the new study offers three examples of the way administrators went about
judging the fairness of actual performance appraisals. One placed about a 75%
reliance on two factors – whether employees who were being appraised were
provided needed information in timely fashion and whether the employees
indicated a liking for the employer. A second virtually ignored these two
considerations and relied about 80% on whether appraisals took account of
employees' views and were based on accurate information. A third paid some heed
to timeliness, accuracy, and workers' liking for the employer but attached the
most importance to whether employees were treated with dignity in the appraisal
process, a factor all but ignored by the other two judges.
Divergent though these approaches were from each other, the study
found a fair degree of consistency in the weight individual judges assigned to different
factors from one appraisal to another. Yet, oddly enough, when asked to rank
factors by importance, their answers were at variance with their actual
judgments. In the words of the study, "Although participants adopted a
consistent judgment policy across different performance-appraisal situations,
they showed little insight into their own judgment policy... [M]ost seem to
have been unaware of the real policies they used."
In short, though experienced in performance appraisals and
consistent in their approach, the judges would have difficulty explaining to
others what that approach was.
The paper reaches these conclusions through an experiment that
hews to actual workplace situations. Fifty-six university administrators (the
experiment's "reporters") were asked to assess an actual performance
appraisal they had recently undergone in terms of eight factors derived from
previous research on organizational justice – whether the performance appraisal
reflected (1) the reporter’s true work effort and (2) true work quality;
whether (3) it took the his or her views and perspectives into account; whether
(4) the appraisal was based on accurate information; whether (5) the reporter
was treated with dignity and respect; whether (6) appraisers refrained from
improper remarks; whether (7) they provided reasonable explanations regarding
the appraisal process; and whether (8) essential information was provided in
good time. Reporters rated each item on a scale of 1 (to a very small extent)
to 7 (to a very large extent), and applied the same scale to two additional
items – (9) whether they were pleased with the appraisal outcome and (10) liked
working for their employer.
These anonymous assessments were then submitted to a group of 49
other university administrators experienced in performance appraisals (the
experiment's "judges"), who were asked to rate each reporter's appraisal
experience on a scale of 0 (totally unfair) to 100 (totally fair). The
importance assigned by each judge to the 10 antecedent factors in the
reporters' surveys was calculated through analysis of the relationship of the
antecedents to judges' fairness ratings over the course of 56 appraisals. In
addition, the judges were asked to provide their own sense of how they ranked
the 10 factors in importance and what percentage each contributed to their
estimates of fairness.
The three antecedent factors rated highest by the judges were
timeliness (factor 8 above), accuracy (4), and quality (2), which had a
cumulative weight of about 42.6 out of 100. The three lowest-weighing factors
were being treated with dignity (5), feeling pleased with result (9), and lack
of improper remarks (6), which had a cumulative weight of only 10.5
More striking was the great disparity among judges on which
factors were important and which were not in achieving fairness. In the words
of the study, "although some antecedents were generally more salient,
there were large individual variations." This despite the fact that
performance appraisal "seems a rather conservative test of agreement on
justice situations, [being] highly standardized and often widely discussed
by employees, thereby making agreement on justice standards more likely than
for many other situations."
Indeed, given the lack of consensus about which factors make for
fairness, simply weighing each factor equally, the study concludes, does
"on average...almost as good a job of explaining global fairness."
The fact "that fairness policies can be so diverse may be
burdensome for...those who conduct performance appraisals," the study concedes.
Still, "focusing on only a few known antecedents of fairness seems risky,
as managers typically do not know which antecedents may be most salient to
which employee. Importantly, managers need to realize that their own
preferences for specific aspects of fairness may not be shared by their
Thus, to the question of whether it is worthwhile for managers to
ask employees which justice considerations are important to them, the
researchers answer with a qualified yes. "Our study suggests that, aside
from the positive effects of being listened to and receiving attention...there
is value in fulfilling on the antecedents that employees think are
important...Managers, however, should act with caution and not take
self-reported policies at face value – e.g., if people tell you that they only
care about three specific antecedents, it may not be wise to therefore assume
that others are unimportant to them."
In conclusion, the authors see the results of their
performance-appraisal experiment as having broader implications for fostering a
sense of fairness among employees. Given the study's findings "that
cognitive resources restrict people's justice-processing," managers should
“be amenable to questions and discussion” and should take care that, "when
it comes to very complex situations with lots of new information (e.g.,
organizational change)... [they recognize] the importance of a) concentrating
on the most salient aspects in initial communication with employees, b) not
providing too much detail to overwhelm the judgment process... and c) providing
time for employees to process details of the change."
The paper, entitled "Justice Judgments:
Individual Self-Insight and Between- and Within-Person Consistency," is
among the research to be published in the spring 2016 issue of Academyof Management Discoveries, a new journal dedicated to
exploring management issues at ground level. The Academy of Management,
with more than 18,000 members in 123 countries, is the largest organization in
the world devoted to management research and teaching. The Academy's other
publications are Academy of Management Journal, Academy
of Management Review, Academy of Management Perspectives, Academy of Management
Learning and Education, and Academy of Management Annals.