IIHHIHNHHWIN!WWII/WillHIHIHHWHHWI

126
162
THS-

   

 

   
     

  

L I B R A R Y
Michigan Stain
University

This is to certify that the

thesis entitled

PERCEPTIONS OF PERFORMANCE APPRAISAL
BY SUPERIOR AND SUBORDINATE

presented by

Harold R. Smith

has been accepted towards fulfillment
of the requirements for

Ph.D. Higher Education

 

 

degree in

 

Date Februa ry 2 , 1978

 

0-7 639

 

PERCEPTIONS OF PERFORMANCE APPRAISAL

BY SUPERIOR AND SUBORDINATEJ

By

Harold R. Smith

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Administration and Higher Education

1978

ABSTRACT

PERCEPTIONS OF PERFORMANCE APPRAISAL
BY SUPERIOR AND SUBORDINATE

By

Harold R. Smith

The Purpose Of The Study

 

The purpose of this study was to substantiate that differences
existed in the perceptions of performance appraisal by superior and sub-
ordinate in regard to purpose, need and method.

Procedure

A questionnaire designed by the researcher to investigate the
perceptions of performance appraisal by superior and subordinate was
administered to the home office employees (N = 565) of an internationally-
known corporation headquartered in the Midwest during the months of Novem-
ber and December 1977. The response rate was sixty-nine percent (69%).

Findings

The results of this study substantiated the position of other
researchers that the need for performance appraisal is seen as more im-
portant by the superior than the subordinate. It also provided support
for studies which indicated that the preferred method for performance
appraisal is perceived differently by superior than subordinate.

Support was not found for writers who take the position that
the purposes of performance appraisal are seen differently by superiors

than subordinates.

Harold R. Smith

The Study results indicated that both superiors and subordin—
ates prefer an appraisal system which provides an Opportunity for the
ratee to rate himself.

Objective appraisal criteria were not seen as important,
especially to subordinates, as might have been expected. Most respon-
dents seemed to favor a system which allowed for at least some subjective
criteria.

Respondents in the study seemed to feel that salary increases
should be closely tied to performance appraisals.

Educational level appeared to have an effect upon an indivi-
dual's perception of performance appraisal. More educated persons
favored more frequent appraisals and gave greater weight to objective
criteria than did less educated persons.

Wbmen indicated a greater preference for an appraisal system
that allowed for ratee input than did men. In ranking performance
appraisal purposes, women also gave greater weight to improving communi-
cation and determining salary than did men, who rated professional growth
higher than either of these items.

Due to limitations of sample composition, the effect of ethni-
city upon the factors studied was not determined. The sample was also
relatively highly educated, with fifty-four percent (54%) having a
college degree and an additional twenty-six percent (26%) having some
college. No attempt was made to examine the effect of occupational type
on the perception of performance appraisal. Age was not found to have a

significant effect on perceptions of performance appraisal in this study.

Ill/I'llil,‘ I

11" I‘ll

To my father,
Dr. J. Harold Smith, whose example, encouragement,

and assistance were of immeasurable value in the
completion of this dissertation.

ii

TABLE OF CONTENTS

Chapter

I.

II.

III.

IV.

INTRODUCTION . . . . . . . . .

Need a o o 0

Summary. 0 o o o o o o o 0
Purpose. . . . . . . . . .
Hypotheses. . . . . . . . .
Overview . . . . . . . . .
Definition of Terms. . . . . .

A REVIEW OF THE LITERATURE . . . .

The Need For Performance Appraisal.
Problems With Perfbrmance Appraisal
What Can Be Done. 0 o o o o 0
Methods of Performance Appraisal .
Other Factors That Affect Performance
Appraisal o o o o o o o o
smary. o o o o o o o o 0

RESEARCH METHODOLOGY AND DESIGN OF THE STUDY

Purpose. 0 o o o o o o o 0
Selection of Research Method. . .

Testing of Hypotheses and Instrument
WSign O O O O O O O O 0
Selection of Sample.

Review System. . . . . . . .
Sample Description . . . . . .
AnaIYSis o o o o o. o o o o
Assumptions . . . . . . . .
Summary. . . . . . . . . .
ANALYSIS OF THE DATA . . . . . .
Response Rate. . . . . . . .
Analysis of Results. . . . . .
Discussion. . . . . . . . .
5mm. 0 o o o o o o o 0

iii

Page

11
11
11
12
13

15

15
21
27
33

47
49

52

52
52

53
56
S7
58
6O
6O
61

62

62
62
69
73

Chapter Page

V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS . . . . . 76

InterUCtion O O O O O O O O O O O O O 76

diapter I O O O O O O O O O O O O O 76

Chapter II. . . . . . . . . . . . . 77

Chapter III 0 o o o o o o o o o o o 78

Chapter IV. 0 o o o o o o o o o o o 79

Conclusions . . . . . . . . . . . . . 79

Inferences and Recommendations . . . . . . . 82

Implications for Future Research . . . . . . 83
APPENDICES

APPENDIX

A. COVER IETTER O O O O O O O O O O O O O O 85

B. QIJESTIONNAEE. O O O O O O O O O O O O O 86

SEIECTED BIBLIOGRAPHY O O O O O O O O O O O O O 88

iv

LIST OF TABLES

Table Page
4.1 RESPONSES TO QUESTION #9 . . . . . . . . . . 63
4.2 RESPONSES TO QUESTION #10 . . . . . . . . . . 64
4.3 RESPONSES TO QUESTION #11 . . . . . . . . . . 65
4.4 RESPONSES TO QUESTION #14 . .. . . . . . . . . 66
4.5 RESPONSES TO QUESTION #12 . . . . . . .1 . . . 68
4.6 RESPONSES ‘IO QUESTION #13 . . . . . . . . . . 69
4.7 RESPONSES TO QUESTION #10 . . . . . . . . . . 70
4.8 RESPONSES TO QUESTION #12 . . . . . . . . . . 72
4.9 RESPONSES TO QUESTION #14 . . . . . . . . . . 73

CHAPTER I

INTRODUCTION

Performance appraisal is a fact of life in any organization,
whether or not it is formalized. Variously termed performance
"appraisal," "review," "evaluation," "assessment," "rating" - perfor-
mance appraisal of some type is an integral part of all organizational
systems and refers to that activity in which a superior evaluates a
subordinate's performance of his job functions. Miner states, "It
should be understood at the outset that the fact of evaluation in some
form is inevitable. All organizations make some effort to determine
whether individual members are contributing to the attainment of objec-
tives."1 Similarly, Kellogg, "All managers make appraisals of other
people and make them frequently."2 Odiorne states that, "In motivating
people to be productive and creative in their work, we cannot rely on
money alone . . . neither fear of economic punishment nor desire for
economic reward can explain the full scope of human motivation."3 ---

thus performance appraisal. Kellogg states it even more strongly, "The

 

1John B. Miner, Personnel and Industrial Relations, (New York:

The Macmillan Co., 1969), p. 180.

2Marion 5. Kellogg, What To Do About Performance Appraisal,

(New York: AMACOM, 1975), Revised Edition, p. 1.

3George S. Odiorne, Personnel Policy: Issues and Practices,
(Columbus, 0.: Chas. E. Merrill Books, 1963), p. 305.

'4

2
question is not whether appraisal is desirable, since it cannot be
avoided."4 Rather, says Kellogg, the questions are: for what purpose,
on what basis, on whose initiative, and with whom should it be dis-
cussed?5
Granted that performance appraisal is inescapable, what is its

purpose? mcGregor's view is that

"Formal performance appraisal plans are designed to meet three

needs, one for the organization and two for the individual: (1)

They provide systematic judgments to back up salary increases,

promotions, transfers and sometimes demotions or terminations.

(2) They are a means of telling a subordinate how he is doing, and

suggesting needed changes in his behavior, attitudes, skills or job

knowledge; they let him know where he stands with the boss. (3)

They also are being increasingly used as a basis for the coaching

and counseling of the individual by the superior."6
More simple and concise is Odiorne's statement that "effective appraisal
of performance and potential of a subordinate by his superior will im-
prove his effectiveness"7 assuming, of course, that most people have a
desire to succeed and will make an effort to improve. Dow Corning
Corporation lists the fellowing objectives for its appraisal system:

1) Recognize individual contribution.

2) Identify individual capabilities.

3) Improve individual performance.

4) Assist in personal development.

5) Prepare people for the professional and managerial positions
that will be needed in the future.

 

4Kellogg, op.cit., p. 4.

5Ibid., p. 4.

6Douglas MCGregor, Leadership And Motivation, (Massachusetts:
M.I.T. Press, 1966), p. 185.

 

7Odiorne, op.cit., p. 304.

3
6) Evaluate fairly, performance as it relates to position expecta-
tions and as it relates to the quality of individual perfor-

mance among the others.8

Need

 

While there is ample documentation indicating the universality
of the existence of performance appraisal and relative unanimity as to
its purpose by management and authorities in the field, there is consid-
erable evidence to indicate that: (1) the purpose is seen differently
by superiors and subordinates, (2) there is disagreement as to its rela-
tive importance, and (3) there is a widespread lack of consensus as to
the form it should take. In short, performance appraisal, when put into
practice, becomes controversial and frequently has an effect upon both
the superior and subordinate that is not only non-productive, but often
is counter-productive.

Dealing first with the question of purpose, Bassett views per-
formance appraisal as primarily serving management's purposes, at least
in the way it is usually conducted.

"Many articles and instruction manuals on the subject discuss in
glowing terms the mutuality of the discussion, the high degree of
agreement and understanding, and other desirable results to be had
from appraisals. In truth, however, a performance appraisal is
essentially the boss's evaluation of a subordinate's work. The
boss spells out what he likes and dislikes about the employee's
performance andé in effect, tells him in what ways he is a good or
poor employee."

"Performance appraisal is a useful method of transmitting information

from manager to employee. There are, however, few, if any, such

 

8Dow Corning, Guide in Reviewing Professional Employees, Nov.

1976, Dow Corning Corp.

 

9Glenn A. Bassett, The New Face of Communication, Second
Edition, (New York: American Management Assoc., 1968), p. 109.

4

discussions in which two points of view are examined. Rather, every-
thing else is put aside by both participants so that the one crucial
point of view - the manager's - can be fully and completely understood."10

This somewhat cynical view, however, does not appear to be
widely shared. Myers' opinion seems to put things in better perspec-
tive. "Like all systems, performance review as a system is at the mercy
of its users. Qualified and committed people can achieve good results
with a poorly designed form and ambiguously defined procedures. Un-
qualified and uncommitted people will cause performance reviews to fail
no matter how sophisticated the system."11

Kellogg believes that "A manager should discard the notion
that his appraisal is an absolute evaluation - rather it's a subjective
judgment based on incomplete information. Subjectivity limits perfor-
mance appraisal but conversely is no different than any other decision
a manager makes."12

White and Barnes point out that some people see performance
appraisal as "something that a supervisor does for the organization to a
subordinate."13 Rieder contends that while superiors may be trying to

use performance appraisal to improve performance, the result may be the

 

lOIbid., p. 115.

11M. Scott Myers, Every Employer A Manager, (New York: McGraw-

Hill Book Co., 1970), p. 69.

12Kellogg, op.cit., p. 61.

138. Frank White, and Louis B. Barnes, "Power Networks in the

Appraisal Process," Harvard Business Review, Vol. 49, May 1971, p. 102.

5
reverse: performance can actually go down or the subordinate may
merely try to emulate his superior.14 Sofer and Tuchman, in a recent
study found that in a comparison of fifty-two (52) superior/subordinate
pairs regarding their respective perceptions of an appraisal interview,
forty-five percent (45%) disagreed as to the main point discussed and
fbrty percent (40%) disagreed as to how the subordinate was rated!15
Richards found in a study of seventy (70) SUpervisors and subordinates
that most subordinates were not anxious to get an evaluation so they
could improve their performance (management's intent); instead, most
saw performance evaluation as a means of obtaining recognition or re-
assurance for the job they were doing.16 Perhaps one of the most strik-
ing examples of the differences in the perceptions of the purposes of
performance appraisal is cited by Kellogg, based upon his experience in
management training at various corporations and locales around the
country. One of his teaching techniques involved giving case histories
of "JOhn JOnes" to management groups. They were divided in half - one
group to be "Jehn Jones", the other his supervisor. Even though both
groups were made up of the same management level people, the very fact
that they were put into the respective roles of supervisor and subor-

dinate brought about a "striking difference in the way the two groups

 

14George A. Rieder, "Performance Review - A Mixed Bag,"

Harvard Business Review, Vol. 51, Ju1y 1973, p. 64.

15C. Sofer and ML Tuchman, "Appraisal Interviews and the

Structure of Colleague Relations," Sociology Review, Vol. 18, Nov. 1970,
pp. 365-91.

 

16Kenneth Richards, "A New Concept in Performance Appraisal,"

appearing in Thomas L. Whisler and Shirley F. Harper, Performance
Appraisal, (New YOrk: Holt Rinehart, & Winston, 1962), p. 560.

 

6

viewed the situation."17 Kellogg fbund that each group was highly
selective in the material they selected to support their positions in
an appraisal interview situation. He found that the fecus for the
groups who played the role of supervisor was to get "Jehn Jones" to do
a better job and was much more "immediately-oriented," while the "JOhn
anes group" tended to focus on some sort of appeasement and was more
future-oriented. Another of his findings was that the "halo effect"
operated strongly, with gray areas seen as all white by the "JOhn JOnes
group" and all black by the "supervisor group". In addition, both
groups appeared "pretty inflexible". Kellogg found these trends present
wherever he used this technique.18 In a study of ninety-two (92) em-
ployees at General Electric, Meyer, Kay and French fbund that the average
self-rating for an employee was at the seventy-seventh (77th) percentile
and only two rated themselves below average. Thus, given a usual spread,
they conclude that for seventy (70) to eighty (80) percent of the employ-
ees, performance appraisal would be a deflating experience.19

The evidence in much of the literature, then, appears to indi—
cate a marked divergence of opinion on the part of the superior and sub-

ordinate as to the purpose of performance appraisal. While the superior

appears to regard performance appraisal primarily as an instrument to

 

17Marion 3. Kellogg, "New Angles In Appraisal," appearing in

Thomas L. Whisler and Shirley F. Harper, Performance Appraisal, (New
Ybrk: Holt Rinehart, & Winston, 1962), p. 90.

 

18mido, p. 91-920

19Herbert H. Meyer, Emmanuel Kay, and John R. P. French, "Split

Roles in Performance Appraisal," Harvard Business Review, Jan.-Feb. 1965,
p. 123.

7
bring about improvement in the job performance of a subordinate, the
subordinate apparently sees performance appraisal as a means of justi-
fication or recognition for the job he is doing.

It is evident, also, that there is widespread disagreement as
to the preferred method for performance appraisal. Even among experts
in the field there is a distinct lack of consensus, for in reviewing
the literature of the past twenty years, one is struck by the fact that
many of the arguments over the relative merits of one method over
another are still going on. Odiorne cites an incident in which he
called together leading experts in the field to develop a program for
performance appraisal. The result, he said, was "chaos - no one could
agree."20

Swerdlow, in a survey of two hundred and five (205) executives,
found that there was no agreement between perception of the most impor-
tant appraisal criteria by the subordinate and the appraisal criteria
actually used by his superior. In fact, among all superior-subordinate
groups tested, no significant agreement was found.21 Similarly, in a
survey of fifteen (15) department chairmen and one hundred seventy-six
(176) faculty in community colleges, Connolly found that chairmen think
that more weight should be given to rank, publication, and research than
do faculty; also, that faculty disagreed with the weighting they think

chairmen give to every one of the twenty-six (26) criteria listed in the

 

20Odiorne, op.cit., p. 307.

21Robert.A.,Swerdlow, An Investigation And Evaluation of
Measures Of Performance Of Executives Relating To Four Types Of Retail
Operations, Doctoral Dissertation, University of Arkansas, 1976.

8

study.22 Blackburn and Clark, comparing correlations of administrator,
colleague, student and self-ratings, found that a professor's percep-
tion of his performance was least correlated with that of his adminis-
trator's assessment of his performance.23 Williams, in a survey of two
hundred and ten (210) directors, managers, and specialists in eight
companies, found that generally managers find appraisal and counseling
much more satisfactory as appraisers than they do as appraisees.24

Thus it clearly appears that superior and subordinate also
have differing opinions as to the methods of performance appraisal, the
criteria used, and the relative weight assigned to each criterion.

Finally, superior and subordinate appear to differ in the rel-
ative importance they assign to performance appraisal and the need for
it, although the evidence in this regard is not as strong as it is for
their differing viewpoints on purpose and method and is, in fact, some-
what conflicting. For example, MCGregor tells of an opinion poll taken
in a company with a well-administered and planned performance appraisal
program in which over ninety percent (90%) of the respondents indicated
that they approved of the idea of performance appraisals, yet forty per-

cent (40%) said they had never been told the results of their appraisals.

 

22Eugene F. Connolly, The Evaluation of Massachusetts Community

College English Teachers: Comparisons of Perceptions of Chairmen And
Faculty, Doctoral Dissertation, Boston College, 1975.

23Robert T. Blackburn, and Mary Jo Clark, "An Assessment of

Faculty Performance: Some Correlates Between Administrator, Colleague,
Student and Self-Ratings," Sociology of Education, Vol. 48, Spring
1975, pp. 242-256.

 

24M. R. Williams, Performance Appraisal In Management, (London:

William Heinemann, Ltd., 1972), p. 89.

9

This, in spite of the fact that over four/fifths (4/5) of them had
signed performance appraisals which were in their files.25 Richards,
in a survey of seventy (70) supervisors and employees, found that all
respondents felt that performance appraisal was necessary. "Despite
misgivings of one kind or another, supervisors and employees still are
"interested in having a performance appraisal program."26

On the other hand, Williams, in his study (mentioned previous-
ly), found that only fifty-six percent (56%) of the respondents felt
performance appraisal was useful, twenty-six percent (26%) did not, and
eighteen percent (18%) were undecided. Van Duren found, in a study of
engineering personnel in the aerospace industry, that the more highly
educated employees who were doing well appeared to be motivated by fac-
tors unrelated to performance appraisal, and that the more senior em-
ployees who had reached a plateau in career growth tended to be below
average in performance and were prone to believe that performance
appraisal is not worthwhile.27

Thompson and Dalton state that the effects of a low perfor-
mance rating result in a decline in morale and loss of work effective-

ness. They indicate further that the lowest-rated employees tend to

stay and that if a high-rated employee's rating is lowered, he gets

 

25M’cGregor, op.cit., p. 186.

26Richards, op.cit., p. 554.

27Allan Van Duren, The Value Of Performance Appraisal In The

Motivation Of Engineering Personnel, Doctoral Dissertation, United
States International University, 1976.

10

discouraged.28 These findings are substantiated by Meyer, Kay and
French, whose studies indicated that criticism had a negative effect
upon the achievement of goals and that praise had little effect, one
way or the other.29 These studies obviously conflict with manage-
ment's position that performance appraisal is needed to improve the
employee's effectiveness.3O Even Odiorne admits, however, that there
is no research evidence to indicate that any appraisal system ever
"developed" anybody.31

Perhaps Kellogg sums it up best. "But if an employee is
appraised by his manager, there is no general way of knowing whether
he feels a need for the appraisal, or whether he trusts the appraiser,
or whether he finds the evidence compelling. He will, after all, make
his personal decision as to whether to take certain actions or change
certain attitudes based on his 2gp appraisal of the total situation."32

The evidence, then, while not compelling, does seem to indi-

cate that the need for performance appraisal is perceived differently

by superior and subordinate.

 

28Paul H. Thompson, and Gene W. Dalton, "Performance

Appraisal: Managers Beware," Harvard Business Review, Vol. 48,
January 1970, pp. 149-57.

 

29Meyer, et.a1., op.cit., p. 123.

30Odiorne, op.cit., p. 306.

3llbid., p. 322.

32Kellogg, op.cit., p. 7.

11
SummaEy
There are thus strong indications by researchers such as
Rieder, Sofer and Tuchman, Kellogg, Meyer, Kay and French, that the
purpose of performance appraisal is seen differently by superiors and
subordinates.
Similarly, evidence provided by Odiorne, Swerdlow, and
Williams, indicates that superior and subordinate do not agree on the
methods of performance appraisal, the criteria used, nor the relative
weight assigned to each criterion.
Finally, studies by Van Duren, Thompson, Dalton, and Meyer,
Kay, and French, appear to indicate that there is similar dissonance in
the opinions of superior and subordinate regarding the need for perfor-
mance appraisal.
ose
The purpose of this study is to substantiate that differences
exist in the perceptions of performance appraisal by superior and sub-
ordinate in regard to purpose, need and method.

Hypotheses

 

Hypothesis 1:

 

The need for performance appraisal will be seen as more important
by supervisors than subordinates.

Hypothesis 2:

 

The preferred method for carrying out a performance appraisal will
be perceived differently by supervisors and subordinates.

A. Supervisors will prefer a method which maximizes their

input.

B. Subordinates will prefer a method which maximizes their

input.

12

C. Supervisors will prefer a method which gives more weight to
subjective criteria.

D. Subordinates will prefer a method which gives more weight
to objective criteria.

Hypothesis 3:

 

,The pggposes of performance appraisal will be perceived differently
by supervisors and subordinates.

A. Supervisors will agree that salary increases should be tied
directly to performance appraisal.

B. Subordinates will feel that salary increases should not be
tied directly to performance appraisal.

C. There will be a significant difference in the rank order of
the purposes of performance appraisal for supervisors and
.subordinates.
Overview

Chapter II contains a review of the literature on performance
appraisal and includes an examination of the need for performance
appraisal, problems associated with it and what can be done to alleviate
them; a review of the types of performance appraisal methods and their
strengths and weaknesses; and finally, a review of other factors that
may affect performance appraisal.

In Chapter III, the design of the study is detailed; this in-
cludes a description of the sample selected, an analysis of the instru-
ment design, a listing of the hypotheses to be tested, a description of
the methods used to analyze the data; the closing portion contains a
summary of the assumptions made in the study.

In Chapter IV, the analysis of the results of the study are
presented. This assessment will determine reasons for acceptance or

rejection of hypotheses; it will also include an evaluation of the sig-

nificance of the findings.

13

Definition of Terms

 

Performance Appraisal.--The placing of a value upon an indi-

 

vidual's performance of his or her job functions relative to certain
criteria, either objective or subjective or both. In the literature,
the following terminology is used interchangeably -- (performance)
appraisal, review, rating, evaluation, assessment, ranking. Currently
the term "review" appears to be gaining some precedence, possibly be-
cause it is less "emotionally loaded", but "appraisal" is still the
most prevalently used term.

Perception.--The process of organizing and interpreting sen-
33

 

sory data by combining them with the results of previous experience.
Objective.--Determined by the features or characteristics of
the object rather than the thoughts or feelings of the evaluator.

Subjective.--Determined by the thoughts or feelings of the

 

evaluator.

Salary Exempt.--Those employees who, under the Fair Labor

 

Standards Act are not eligible for overtime pay for overtime work.
Generally they are college graduates who hold a professional or super-
visory position or are involved primarily or substantially in making
policy decisions or using their professional training.

Salary Non-Exempt.--Those employees who, under the Fair Labor
Standards Act are eligible for overtime pay. They are usually non-

professional, non-college-educated employees.

 

33Floyd L. Ruch, Psychology and Life, Fourth Edition, (New-

York: Scott, Foresman & Co., 1953), p. 241.

 

l4

Subordinate.--A person who is subject to the power or author-

 

ity of another.
Spperior.--A person who, by virtue of rank, position or
authority, directs the actions of others.

Supervisor.--A person who oversees and directs the work of

 

others.

XEEE§,--"Values do not consist in 'desires' but rather in the
desirable . . . they are abstract standards that transcend the impulses
of the moment . . ."34 "A cultural value may be defined as a widely
held belief or sentiment that some activities, relationships, feelings,
or goals are important to the community's [or culture's] identity or
well-being."3S

Common Performance Appraisal Errors:

 

Halo Effect.--The tendency, in an appraisal situation,

 

to rate an individual in all areas based upon his rating in one or two
considered most important.

Central Tendency Error.--The tendency to rate all em-

 

ployees within a narrow range.

Recency Error.--The tendency to rate a person primarily
based upon a recent major incident rather than for the whole time period
covered.

Personal Bias Error.--The tendency to rate a person on

non-objective factors reflecting the personal biases of the rater.

 

34Clyde Kluckhohn, Culture and Behavior, (New York: The Free
Press of Glencoe, 1962), p. 289.

35Leonard Brown and Philip Selznick, Sociology, Fourth

Edition, (New York: Harper & Row, 1968), p. 54.

CHAPTER II

A REVIEW OF THE LITERATURE

A review of the literature on the subject of performance
appraisal leads one to the conclusion that performance appraisal has
generated considerable: (a) controversy, (b) research, (c) development
work and (d) extensive literature.

Considering the amount of writing and research on the subject
that has taken place for the past twenty-odd years, it is remarkable how
little consensus exists as to the purpose, method and even the value of
performance appraisal.' Seemingly, for every opinion or study support-
ing one point of view, there is another opinion or study supporting
another. Confronted with a bewildering array of seemingly "expert
opinion", a student of performance appraisal is left with the view that
very few conclusive or definitive positions can be taken on the subject.

The Need For Performance Appraisal

 

Consider first of all the rather elemental question of whether
performance appraisal has value - should it even be done? James Conant
has a rather strong opinion: "There seems to be no documentation
supporting the proposition that performance appraisals work. They have

evolved as a standard practice and are perpetuated on a 'faith' basis."36

 

36James C. Conant, "The Performance Appraisal," Business

Horizons, vol. XVI, No. 3, June 1973, p. 77.

15

16
He is supported by Meyer. "In theory, the performance appraisal pro-
gram appears to be indisputably sound and logical. Yet the program has
proved to be an enigma to both personnel experts and line managers."
He points out that the rating process and especially the feedback inter-
view have presented almost insoluble problems. "Objective evidence has
shown that appraisal interviews seldom have the positive effect attri-
buted to them." Some research indicates that they may do more harm than
good.37 Among these studies is one in which Meyer, himself, participat-
ed. He and his colleagues found the following: (a) criticism had a
negative effect on the achievement of goals, (b) praise had little effect,
one way or the other, (c) the average subordinate reacted defensively to
criticism, (d) defenSiveness as a result of criticism resulted in inferior
performance, (e) criticism was more disruptive to those already low in
self-esteem (those that needed help the most). They add that the super-
visor often finds himself defending the rating more than uSing it as a
tool to improve performance.38 A

Oberg substantiates that performance appraisal can boomerang

and result in worse performance,39 as does Rieder.4O Patz points out

 

37Herbert H. Meyer, "The Annual Performance Review Discussion -
Making It Constructive," Personnel Journal, Vol. 56, October 1977,
p. 508. '

38Herbert H. Meyer; Emmanuel Kay; and John R. P. French,

"Split Roles In Performance Appraisal," Harvard Business Review,
January-February 1965, p. 123.

 

39Winston Oberg, "Make Performance Appraisal Relevant," Harvard

Business Review, Vol. 50, January 1972, p. 62.

40George A. Rieder, "Performance Review - A Mixed Bag," Harvard
Business Review, Vol. 51, July 1973, p. 63.

 

17

that the interpersonal part of performance appraisal, the appraisal
interview, is difficult and may discourage the employee.41 However,
none of the writers listed in this paragraph suggest abandoning per-
formance appraisal, but rather each offers suggestions to make it more
effective.

Sofer and Tuchman, however, found, in a study of fifty-two
(52) superior/subordinate pairs, that there was substantial disagree-
ment regarding issues discussed in an appraisal interview: among these
were the main point discussed, how the employee was rated, whether any-
thing was said on a given issue, whether the employee received criti-
cism, and whether the employee's strengths and weaknesses were discussed.
In general, they found that the superior said he had discussed an issue
while the subordinate said he had not. Sofer and Tuchman believe that
due to the relative roles of the participants, the appraisal interview
is subject to pressures that force apart the recollections and interpre-
tations of the participants. They openly wonder why management persists
in appraisal interviews when they appear to be so ineffective, "to re-
assure stockholders that they are in control of human as well as material
resources?"42

Meyer and Conant both see the crucial problem as being the fact
that a person's self-esteem is inextricably intertwined with his occupa-

tion. In a very perceptive article, Conant states, [there's a real

 

41Alan Patz, "Performance Appraisal, Useful But Still Resisted,"

Harvard Business Review, Vol. 53, May 1975, p. 77.

42C. Sofer and M. Tuchman, "Appraisal Interviews and the Struc-
ture of Colleague Relations," Sociology Review, Vol. 18, November 1970,
pp. 365-91.

 

l8
dilemma for managers] --- "how to present corrective recommendations
without adversely affecting the man's drive."43
He adds:
"Even if the appraisal is conducted objectively by the manager,
the most accurate assessment is seldom handled candidly by the
employee . . . In my experience, criticism often points out prob-
lems with which the person has been contending much of his life.
He is usually painfully aware of his deficits, and he does not
appreciate having them brought to his attention again . . . His
feelings of inadequacy often are heightened by the appraisal
and significantly diminish his self-esteem --- the very element
most necessary to effectively deal with the problem."44
Meyer confirms the often negative side-effect on the employee's
occupational self-esteem, adding that the employee is more likely to
react defensively than constructively to suggestions for an improvement
in performance. He adds that research indicates that people usually
handle threats to self-esteem in unconstructive ways: either by question-
ing the measurement criteria or minimizing its importance or by demeaning
the source (evaluator).45
Conant cites similar reactions: (a) the employee resolves to
change, does so and shortly after backslides, (b) he goes overboard the
opposite way (exaggerates), (c) he does nothing, but worries and sometimes
intensifies the pattern, or (d) he withdraws. "It is characteristic for
the beleagured individual to regress by exaggerating (his) adjustment pro-

cess." What follows, says Conant, is a vicious circle. The manager be-

comes increasingly frustrated and emphasizes the employee's deficiencies,

 

43Conant, op.cit., p. 73.

44Conant, op.cit., p. 75.

45Meyer, "Annual Performance . . . ," p. 508.

19

thus intensifying the employee's anxieties. As the manager becomes a
threat, he can no longer be a resource. And so, Conant concludes, per-
formance appraisal never works when you need it --- to handle a tough
problem.46

Despite the evidence that performance appraisal apparently
does not work, or at least, does not work like it is supposed to, it
continues to be used by most companies. Locher and Teel, in a recent
survey of six hundred and ninety-six (696) organizations belonging to
the Personnel and Industrial Relations Association of Los Angeles, found
that eighty-nine percent (89%) of the organizations responding had a
formal performance appraisal program. Although only thirty-one percent
(31%) responded to their survey they felt that the five-county area they
surveyed was one of the largest population and industrial centers in the
United States and they believed that this area was likely representative
of general practice across the country.47 And, as Kellogg says, "The
question is not whether appraisal is desirable, since it cannot be avoid-
ed."48

Others have pointed out that the employee, as well as manage-

ment, desires performance appraisal, though, as pointed out in Chapter I,

 

46Conant, op.cit., pp. 75-77.

47Alan H. Locher and Kenneth S. Teel, "Performance Appraisal -

A Survey of Current Practices," Personnel Journal, Vol. 56, May 1977,
p. 245.

 

48Marion S. Kellogg, What To Do About Performance Appraisal.

(New York: AMACOM, 1975), p. 4.

20
. 49
perhaps for different purposes.

Theorists, too, make a seemingly logical case for performance

appraisal. Two pertinent statements follow:
"The theory of appraisals is that the motivation of employees in
administrative organizations can be achieved in part by doing two
extremely basic things:

1) The superior lets his subordinates know what is expected
of them -- what constitutes good performance and unsatis-
factory performance.

2) He uses these standards of good and poor performance subse-
quently to let the employee know how well - or poorly - he
has performed over a specified period just completed.50

Williams states, " [Performance appraisal's] significance is

underlined by the comparatively higher frustration, dissatisfaction,
and staff turnover where it is not carried out, or, at best, done
badly".51

It appears, then, that performance appraisal is a procedure

that almost everyone, including superiors, subordinates, and management
theorists think we need; however, they agree that such appraisal does
not accomplish what it is claimed to accomplish or, at best, does it

rather badly. If, one wonders, almost "everyone" says we need it and

"everyone" uses it, why does it not work?

 

49Kenneth Richards, "A New Concept In Performance Appraisal,"

appearing in Thomas L. Whisler and Shirley F. Harper, Performance
Appraisal. (New York: Holt Rinehart & Winston, 1962), p. 554.

M. R. Williams, Performance Appraisal in Management. (London: William
Heineman LTD., 1972), p. 81.

Douglas McGregor, Leadership And Motivation. (Massachusetts: M.I.T.
Press, 1966), p. 186.

 

50George S. Odiorne, Personnel Policy: Issues and Practices.

(Columbus, 0.: Chas. E. Merrill Books, 1963), p. 306.

51Williams, op.cit., p. 95.

21

Problems With Performance Appraisal

 

Many, if not most, of the problems with performance appraisal
have to do with the "human factor". As indicated earlier, one of the
major difficulties stems from the fact that performance appraisal deals
with an evaluation of an individual's job performance. Typically a
person has a great deal of himself invested in his job. For many people,
their jobs are of major, even prime, importance in their lives. Quite
naturally, therefore, any evaluation of their job performance is bound
to be an emotionally difficult and, at times, even threatening, exper-
ience.

A major problem, as was pointed out so well by Conant, is that
performance appraisal fails where it is needed most --- in dealing with
an employee whose performance is sub-standard. Van Duren indicates that
senior employees in his study who tended to be below average in perfor-
mance felt that performance appraisal wasn't worthwhile.52 And as
Meyer,53 Oberg,S4 Rieder,SS and PatzS6 point out, a critical appraisal
frequently results in a decline in performance.

Another problem originates with the superior who does the

rating. Frequently the rater has had little training in rating and has

 

52Allan Van Duren, The Value Of Performance Appraisal In The

Motivation Of Engineering Personnel. Doctoral Dissertation, United
States International University, 1976.

53Meyer, op.cit., p. 508.

54Oberg, op.cit., p. 62.

55Rieder, op.cit., p. 63.

56Patz, op.cit., p. 77.

22
little interest in it, considering it a "necessary evil". Oberg lists
some of these obstacles:

1) Performance appraisal programs demand too much from supervi-
sors - especially if they are expected to evaluate twenty to
thirty subordinates.

2) Standards and ratings vary widely among raters and often
unfairly.

3) Personal values and bias can replace organizational standards.

4) Validity may be reduced by the supervisor's resistance to per-
formance appraisal (avoid problems, rate everyone average to
above average).

Patz lists additional difficulties; he says that because un-
favorable performance appraisals reflect on the superior as well as the
subordinate and excessively favorable performance appraisals may cause
problems with peers or lead to the loss of the best employees through
intra-company transfer, there is a tendency to "strike a happy middle"
in ratings. He adds that most supervisors talk too much in performance
appraisals and listen too little.58

Grey and Kipnis found that the greater the proportion of non-
compliant workers in a unit, the more favorable would be the supervi-
sor's judgments of his/her compliant workers; and conversely, a few non-
compliant workers in a context of compliance will be evaluated more
harshly.59

Odiorne points out that the appraisal process is "just as much

 

57Oberg, op.cit., p. 63

58Patz, op.cit., p. 76.

59Ronald J. Grey and David Kipnis, "Untangling The Performance

Appraisal Dilemma: The Influence Of Perceived Organizational Context On
The Evaluative Process," Journal of Applied Psychology, Vol. 61, JUne
1976, pp. 329-35.

23
a reflection of the manager making the appraisal as it is of the person
being appraised."6O He lists reasons why people can be down-rated by a
supervisor:
l) The boss is a perfectionist.
2) The ratee disagrees with the boss too often.-
3) The ratee is a non-conformist.
4) The ratee is part of an overall weak group.
5) The ratee hangs around with a "frivolous", etc., crowd.
6) The ratee "goofed" in a big way recently.
7) The ratee's traits rather than his performance are rated.
8) The boss thinks he did a better job than the ratee when he
held the ratee's position.61
On the other hand, the hypothesis that superiors would rate
more highly subordinates with similar personality and biographical traits
has been rejected in three separate studies.62’ 63’ 64
Locher and Teel, in their study, found that about fifty-five
percent (55%) of the appraisers had never received training in rating
and that another twenty percent (20%) had only received initial in-

structions. They also found that almost ninety-nine percent (99%) of

appraisals were done by the immediate supervisor.65

 

60Odiorne, op.cit., p. 306.

61Odiorne, op.cit., p. 312.

62Basil Irvin Summers, Exploratory Study: Comparison of

'Cognitive Spyles' To Employee Evaluation And Employee Satisfaction.
Doctoral Dissertation, Wayne State university, 1971.

63Norman Andrew Huber, Superior-Subordinate SimilaritylAPer-

formance Evaluation And Job Satisfaction. Doctoral Dissertation, wayne
State university, 1970.

64James Tester.Quinn, Performance.Appraisals: The Relation

Between Ratings And Selected Characteristics Of_The Rating Dyads.
Doctoral Dissertation, Indiana University, 1967.

6SLocher and Teel, op.cit., p. 247.

24

An additional source of difficulty is that most performance
appraisals tend to be annual (fifty-two percent (52%) in Locher and
Teel's study) or semi-annual (twenty-four percent (24%) in the same
study).66 Meyer describes the annual appraisal as a psychologically
unsound procedure. He points out that it's a well-established fact
that feedback is much more effective when it's associated with the act
it's meant to correct.67 unquestionably, too, an annual or semi-annual
review is bound to increase recency error.

Yet another obstacle results from a lack of clarity of pur-
pose. Confusion results when performance appraisal is used to serve
too many purposes. What is its purpose? When one attempts to answer
that question, he is met with a considerable variety and number of
responses.

Oberg lists the following:

1) To help supervisors observe subordinates.more closely and do a
better coaching job.

2) To motivate employees by giving them feedback.

3) To provide back-up data for management re: raises, transfers,
dismissals, etc.

4) To identify employees with promotion potential and pinpoint
development needs. _

5) To establish a research and reference base for personnel
decisions.68

Levinson simply states that the purposes are: (a) to provide
feedback to the employee, (b) to serve as a basis for modifying or

changing work habits, and (c) to provide judgment for compensation and

 

66Ibid., p. 247.

67Meyer, op.cit., p. 508.

68Oberg, op.cit., p. 61.

25

future job assignments.69 Morrisey cautions that performance appraisal
should be most closely related to employee development and not be done
solely at times of salary or promotion consideration.70

In their survey, Locher and Teel inquired as to the primary
purpose to which performance appraisal was put by the company. Seventy-
one percent (71%) said for compensation, fifty-five percent (55%) listed
performance improvement, twenty-nine percent (29%) feedback, twenty-five
percent (25%) promotion, and eleven percent (11%) documentation.71
Kellogg believes that there should be four distinct types of appraisal:
coaching, where the manager attempts to determine how to guide the em-
ployee to improve his performance; career guidance, where he appraises
the employee's overall abilities in order to counsel him regarding his
career; salary, where he assesses the employee's work performance to
determine salary adjustments; and promotion, where he assesses work per-
formance to determine promotion possibilities. Kellogg believes that
these should be kept separate.72 Richards contends that performance
73

appraisal should be a supervisory rather than a management tool.

Closely related to difficulties involving purpose is the

 

69Harry Levinson, "Appraisal Of What Performance," Harvard

Business Review, Vol. 54, July 1976, p. 30.

70George L. Morrisey, Appraisal And Development Through

Opjectives And Results. (Reading, Mass.: Addison-Wesley Publishing
Co., 1972), p. 7.

71Locher and Teel, op.cit., p. 246.

72Kellogg, op.cit., p. 8.

73Richards, op.cit., p. 554.

26
tendency for some companies to attempt to use one or two appraisal forms
to fit all needs and all employees. Sloan and Johnson caution that it
is impossible to adequately evaluate professional employees and, for
example, assembly line workers, in the same manner.74 Thompson and
Dalton caution against the "temptation to devise one grand appraisal
system to serve all management needs (assignments, promotions, raises,
1ayoffs)."75 Rieder cautions against the development of an "eclectic
hodgepodge," a performance appraisal system with unclear objectives,
mixing salary administration, coaching, counseling, and promotion.76
Odiorne lends support, "Perhaps the great flaw in appraisal systems is
that we have been looking too hard for a single system which will apply
to all men".77 Morrisey doubts that a common appraisal form can be
developed for all employees and jobs in a company, but he believes that
all can and should share a common philosophy.78 Salton believes that a
serious problem with performance appraisal is that of standardized for-
79

mats which don't meet the job or evaluator's needs.

Patz lists what he terms collection and analysis obstacles.

 

74Stanley Sloan and Alton C. Johnson, "New Context of Personnel

Appraisal," Harvard Business Review, vol. 46, November 1968, p. 16.

75Thompson and Dalton, op.cit., p. 57.

76Rieder, op.cit., p. 62.

77Odiorne, op.cit., p. 324.

78Morrisey, op.cit., p. 6.

79Gary J. Salton, "VARIMAT: Variable Format Performance

Appraisal," Personnel Administrator, Vol. 22, JUne 1977, p. 54.

27

under collection obstacles he states that performance appraisal data
is often too qualitative and where quantitative data exists, the rating
scales are often too narrow, making it too difficult to discriminate
between employees. He also feels that data collection is often done in
a haphazard fashion. The second difficulty is dealing with and mean-
ingfully analyzing the data once it's collected.80

Williams sums it up rather nicely: "The main impression from
the research . . . was that performance appraisal is both more impor-

81

tant and more complex than was previously believed."

What Can Be Done

 

In spite of problems resulting from human inadequacies, lack
of clear purpose and cOllection and analysis obstacles, few people advo-
cate dispensing with performance appraisal entirely. Much of the liter-
ature, on the contrary, deals with suggestions to make it work or to
make it work more effectively. Thompson and Dalton offer the following
suggestions: 1) In providing feedback to an employee, use many kinds
and avoid comparing him to others, and 2) Keep the performance appraisal
system Open and future-oriented.82 Sloan and Johnson emphasize that per-
formance should be evaluated in relation to established goals and objec-
tives as opposed to "desirable traits". They suggest that those who

evaluate and develop an organization's performance appraisal program ask

 

8OPatz,.op.cit., p. 75.

81Williams, op.cit., p. 95.

82Thompson and Dalton, op.cit., pp. 56-7.

28
the following questions:

1) Is performance appraisal to be primarily a planning or a
control tool?

2) What and how accurate are my assumptions regarding the behavior
and motivation of this organization's employees?

3) Are the purposes of the current performance appraisal system
clearly specified and does the system meet those purposes?

4) Are there other, better methods?

5) What is the relationship of the individual's performance to
the organization's goals?

Rieder presents the following remedies:

1) Superior and subordinate should develop a "partner" approach,
with joint planning and assessment.

2) The individual should be considered. Not everyone wants to
pay the price of major job growth. There should be an allow-
ance for creativity; the individual must achieve self-satis-
faction.

3) Build on the employee's strengths, bypass weaknesses.

4) There must be managerial support year-'round, not just at
appraisal time.

5) The system should be flexible -- the workforce needs to be a
mixture, not homogeneous.

6) Feedback should be quantitative.

7) Management should commit itself to action - don't put off
answers to questions or acting on suggestions.84

Patz offers four rules: 1) Keep it simple - scales should be

 

fairly wide; you can use performance appraisal to determine who gets a

raise, but not how much. 2) Keep it separate - separate from MBO (man-

 

agement by objectives) programs, especially for promotions. For example,

the salesman might not be the best sales manager. 3) Keep it contained -

 

tailor it to the job and limit its purpose to appraisal. 4) Keep it
participative - there should be two-way communication, including a self-
assessment by the subordinate, and the superior should seek out what he

can do to assist the subordinate.85

 

83Sloan and Johnson, op.cit., p. 20

84Rieder, op.cit., pp. 66-7.

85Patz, 0p.cit., p. 80.

29
There are a number of empirical studies which support this

latter point --- active participation by the subordinate in the appraisal
process. Greller found that subordinates were more satisfied with ap-
praisal interviews where they were invited to participate.86
French, et.a1., conducted a study of ninety-two (92) managers which
involved manipulating the amount of participation by the appraisee in
setting goals and planning methods for achieving the goals in improving
performance. They followed up with the superior and subordinate twelve
weeks later and found:

1) Higher participation was associated with greater self-actualiza-

tion.

2) Participation produced improvement in superior/subordinate rela-
tions.

3) Formulation of criticisms of past performance into specific
goals for future improvement had a very substantial effect on
performance improvement.

4) While increases in participation improve the relationship be-
tween superior and subordinate, decreases do not have the ex-
pected undesirable results.

They also found that participation by the subordinate was
affected by the type of superior: if the superior was usually participa-
tive and supportive, increased participation had a positive effect, but
if the superior was usually a threatening and critical type, increased
participation had a negative effect [possibly because the subordinate was
suspicious? . The study also indicated that subordinates who are more

independent function better with more participation, while those who are

less independent function better in less participative appraisal

 

86Martin M. Greller, "Subordinate Participation and Reactions

to the Appraisal Interview," Journal of Applied Psychology, Vol. 60,
October 1975, pp. 544-9.

30

situations.87

Williams, too, supports a participative approach. "Joint
target setting, while no panacea, does appear to provide specific focal
points or common frames of reference which enable superiors and subor-
dinates to concentrate both their thinking and mutual activities on the
right priorities and most critical issues."88

One study contrasts with this view; however, by the authors'
admission, the situation may have affected the results. The study con-
sisted of appraisal interviews of sixteen (16) student teachers for a
micro-teaching session. In this situation, the findings indicated that
non-participative performance appraisal interviews were more effective
than participative interviews in changing behavior and in rater satis-
faction. The authors conclude that there may be a difference in train-
ing and job situations, thus certain conditions might warrant a non-
participative approach.89

Baird reports an interesting study regarding the inter-rela-
tionship of self-ratings and self-esteem. He found the following: (a)
subordinates with high self-esteem will agree with high ratings by their
superior and disagree with low ratings, (b) subordinates with low self-

esteem will disagree with high ratings by their superior and agree with

 

87John R. P. French; Emmanuel Kay; and Herbert H. Meyer,

"Participation and the Appraisal System," Human Relations, Vol. 19,
February 1966, pp. 3-20.

 

88Williams, op.cit., p. 90.

89Joseph M. Hillery and Kenneth N. wexley, "Participation
Effects In Appraisal Interviews Conducted In A Training Situation,"
Journal of Applied Psychology, Vol. 59, April 1974, pp. 168-71.

31
low ratings, and (c) subordinates with high self-esteem will report low
satisfaction with superiors who rate them low. He concludes that self-
ratings are not only performance measures but indicative of an indivi-
dual's self-image. He believes also that for the high self-esteem indi-

vidual, job performance is a source of satisfaction and he is thus un-

willing to give himself a low rating.90

Meyer is another writer who stresses the need for a partici-
pative approach to performance appraisal. He lists six ways to improve
a performance appraisal interview:

1) Minimize authoritarian interaction -- participate as equals.

2) Both superior and subordinate should prepare in advance for
the discussion.

3) The subordinate should make his presentation first so he is
not put on the defensive.

4) Threats to individual self-esteem should be avoided. The
interaction should be problem rather than person-oriented.

5) Avoid a "report card" type of rating form.

6) Don't try to cover all aspects of the job in a single interview.

He suggests the following format for the interview:

A. Introduction - discuss purpose. ’

B. Obtain the employee's view.

C. Present the supervisor's view, avoiding comparisons with other
employees.

D. Discuss behaviors desirable to continue, one or two only.

E. Discuss opportunities for improvement, again one or two items,
work-related.

F. Shape a performance improvement plan, developed by the employee.

G. Discuss future opportunities -- advancement, salary, warning for
poor performer. I

H. Ask fgp questions and close on a constructive and encouraging
note.

 

90Lloyd 5. Baird, "Self and Superior Ratings of Performance:

As Related to Self-Esteem and Satisfaction with Supervision," Academy of
Management Journal, Vol. 20, No. 2, June 1977, pp. 291-300.

 

91Meyer, "Annual Performance . . . ," op.cit., pp. 509-10.

32

In developing a performance improvement plan, Kellogg offers
this approach:

1) Make plans specific and concrete.

2) Avoid suggestions involving the changing of personal character-
istics and traits. Describe behavior that has hurt the indivi-
dual and suggest alternative behavior.

3) Concentrate on growth opportunities within the present position.

4) Limit growth plans to a few important items. 92

5) Focus plans on specifics rather than general improvement.

Ford and Jennings agree with Meyer's earlier Statement that an
annual performance appraisal is illogical and ineffective. They feel
instead, that performance appraisal should be done on a bi-monthly basis.
They believe that this approach offers the following advantages:

1) The immediate attention of the employee is brought to problems;
the evaluation is also likely to be more honest.

2) Review time is reduced, there is less motivation to bicker over
a point.

3) Halo error is substantially reduced; accuracy is improved.

4) Recency error is also reduced.

5) The superior can provide more effective guidance.

6) The organization has current and readily-available data to
respond to any EEO charges.93

There is some evidence that training for raters may have some
beneficial effects if management is willing to invest the time and money.
Borman, in a somewhat inconclusive study, found that halo error was re-
duced by training, though still present, but that inter-rater reliability

worsened, possibly due to greater distinctions in the raters' minds among

performance categories.94 In a somewhat more definitive study, Latham,

 

92Kellogg, "New Angles . . . ," op.cit., p. 88.

93Robert C. Ford and Kenneth ML Jennings, "How To Make Perfor-

mance Appraisals More Effective," Personnel, Vol. 54, March-April 1977,
Pp. 51-6.

94walter C. Borman, "Effects of Instructions to Avoid Halo
Error on Reliability and Validity of Performance Evaluation Ratings,"
Journal of Applied Psychology, Vol. 60, October 1975, pp. 556-60.

33

et.a1., compared the effects of training on sixty (60) managers who
were randomly assigned to a performance appraisal workshop, to a group
discussion on performance appraisal, and to a control group. After six
(6) months, each group was asked to rate persons who appeared on video
tape. The authors found that the control group committed similarity,
contrast, and halo errors; the discussion group committed impression
errors, and the workshop group committed no errors. The workshop method
was the most effective in training but it was also the most expensive

95

and time-consuming.

Methods Of Performance Appraisal

 

As mentioned earlier, there are nearly as many methods for
performance appraisal as there are writers in the field; consequently
the literature is filled with studies supporting or criticizing these
various approaches. In this section, methods will be classified by type,
and then studies and opinions will be reviewed regarding the strengths
and weaknesses of each.

Morrisey provides some general observations regarding subjec-
tive and objective measures. While he concedes that performance apprai-
sals should concentrate mostly on objective measures, he believes that
subjective measures must also be considered. Though he grants that a
person may perform excellently on the job by objective standards, he may
fail as a superior, subordinate, or co-worker. Thus he feels subjective
analysis is needed but must be directed to a person's behavior rather

than to what he is. Additionally, he states, subjective judgments must

 

95Gary P. Latham; Kenneth N. Wexley; Elliott D. Pursell,

"Training Managers To Minimize Rating Errors In The Observation Of
Behavior," Journal of Applied Psyghology, Vol. 60, Oct. 1975, pp. 550-5.

34
be supported by objective data.96
He lists the problems with subjective measures as follows:

1) They evaluate more what the individual is and knows than what
he does.

2) subjective characteristics do not have mutually considered and
reliable definitions for all who must relate to them.

3) It is hard to assess their relative importance to job perfor-
mance or to determine whether they actually correlate to job
performance.

4) Inter-rater and same-rater reliability of ratings on these
characteristics is low.97

McAfee and Green list ten methods of performance appraisal:

1) Employee comparison - Superior ranks subordinates according to
his estimate of their worth to the organization.

2) NBC - Superior and subordinate set goals; superior evaluates
Efibordinate in regard to goal accomplishment.

3) Direct indexes - Subordinates are compared by superior in terms
of objective measures.

4) weighted checklist - Superiors are given a list of statements
and askedsto-check.those that describe a subordinate's behavior.

5) Forced-choice checklist - Superior is forced to choose between
several equally favorable or unfavorable statements which most/
least describe a subordinate's behavior.

6) Essay - Superior is asked to write an essay describing a subor-
dinate's performance.

7) Rating scale: single word anchor - Subordinate is rated on a
number of traits or dimensions, each on a scale (excellent,
good, fair, poor). No definitions for anchors.

8) Rating scale: short phrase anchor - Same as above, except short
phrases placed on scale.

9) Rating scale: _paragraph anchors - Same as above, except para-
graph anchors are placed on scale.

10) Rating scale: behaviorally anchored - Scale anchors consist of
specific job behaviors.98

 

 

 

Oberg comes up with a somewhat different list and critiques

 

96Morrisey, op.cit., p. 5.

97Ibid., pp. 3-4.

98Bruce McAfee and Blake Green, "Selecting A Performance

Appraisal Method," Personnel Administrator, Vol. 22, June 1977, pp. 61-62.

35

each method:

1)

2)

3)

4)

5)

6)

7)

8)

Essay - Superior writes paragraph on strengths and weaknesses

of subordinate. Strengths: easy to construct. weaknesses:
too variable in length and content. Very difficult to compare

ratings. - .
Graphic rating_scale - Subordinate's traits are rated on scale.

 

Most widely used method. Can be forced choice but regular
scale is cheaper to devise, more acceptable to raters and just
as valid in practice. Strengths: easy to use, more consistent
and reliable than essay. Weakness: lacks depth.

Field review - Member of central office meets with raters in

 

groups and goes over employees' ratings. Strengths: more fair
and valid than individual rating. weakness: very time-consum-
ing.

Forced-choice rating - Superior chooses between groups of state-

 

ments that best fit subordinate as opposed to those that least
fit. Statements are then weighted and scored. Strength: rater
doesn't know rating, may be more objective. weaknesses: diffi-
cult to develop, raters and ratees usually both dislike.
Critical incident appraisal - Superior keeps record of positive
and negative employee behavior. Strengths: specific and on-
going. Weaknesses:r very time-consuming, superior sets standards.
Management by objectives - Superior and subordinate set goals;
superior rates subordinate on goal achievement. Strength: con-
centrates on objectives and achievement rather than traits and
behavior. weaknesses: at lower levels, many employees don't
want to set goals or be involved with self-direction. Sometimes
management imposes its own goals.
work standards approach - Organization sets work standards for
each job, employee evaluated accordingly. Strength: if stan-
dards are fair and realistic, method is relatively objective.
Weaknesses: people are evaluated on different standards, job to
job. Promotions and raises are thus hard to determine.
Ranking methods
A. Alternation - From list of names, rater selects "most
valuable" and "least valuable" employee and then fills
in between with other employees in rank order.
B. Paired-comparison - Compare each individual with every
other individual according to Specific criteria, giving
a tally to the higher ranked. Person with most tallies
is most valuable, least tallies is least valuable.
Strength: good way of determining order of merit for
salary increases. weaknesses: cumbersome and time-
consuming§ Hard to compare persons in one group with
another.

In considering the value of each of these for specific purposes,

 

99Oberg, op.cit., pp. 65-6.

36
Oberg rates them as follows:
1) For observing and coaching employees - critical incident.
2) For motivating and giving feedback - MBO and critical incident.
3) For raises, transfers, dismissals - graphic or graphic/essay.
combination.
4) For identifying promotion potential and development needs -
graphic/essay combination.
5) For establishing a research and reference base for personnel
decisions - graphic/essay plus rankings.100
Locher and Teel found the following use pattern for the cor-
porations in their survey: rating scale - fifty-six point seven percent
(56.7%), essay - twenty-four point nine percent (24.9%), MBO - twelve
point seven percent (12.7%), and other - five point seven percent
(5.7%).101
The literature contains a considerable number of articles and
studies criticizing or advocating a given method. Among these, Thompson
and Dalton are critical of any type of peer comparison rating in which
an employee is ranked among his peers. They call this a "zero-sum
game", since for any person to gain, another must lose. If five men are
above average in a ten-man department, then five must be below average.
Often, if an individual improves, so do those above him, so his relative
position stays the same. If they don't improve and he passes them, they
lose in ranking and as has been shown in several studies, a decline in
rating usually results in a decline in morale and loss of work effective-
ness. Often, too, the lowest-rated employees tend to stay, possibly
fearing that they won't be able to get another job.

In practice, say the authors, in order to keep younger men with

 

l001mm, p. 67.

101Locher and Teel, op.cit., p. 247.

37

the company, it has been necessary to "steal" points from older men.
They also noted that supervisors tended to act toward employees on the
basis of their ranking. Another problem is that zero-sum schemes force
a ranking of various tasks within the company. "Is this realistic?" they
ask. All jobs contribute but in various ways. Far better, they feel,
are approaches which evaluate employees' performance against pre-estab-
lished objectives, for example, MBO.102

Patz, however, states that most companies do not think that MBO
can be used to supplant performance appraisal, but should be used with
it. Patz contends that performance appraisal does not always pay off for
the company because it's too individualized. He sees it as more of a
planning than an assessing tool, and he believes that its greatest weak-
ness as an appraisal instrument is that it fails to tell the employee
what behavior he must adopt or modify to improve results.103

Beer and Ruh, too, feel that MBO has weaknesses as an appraisal
tool. First, they believe that it focuses excessively on task results,
and secondly, it does not help the manager to observe and evaluate the
behavior of subordinates, a practice which is needed to assess promotion-
al possibilities or to help improve performance.

They advocate instead a three-part system as developed by Corn-
ing Glass called the performance management system (PMS). PMS is distin-
guished from other performance appraisal systems by: (a) recognition of

the triple role played by superiors with subordinates (manager, judge,

and helper), (b) emphasis on both development and evaluation, (c) use of

 

102Thompson and Dalton, op.cit., pp. 49-55.

103Patz, op.cit., p. 75.

38
a profile displaying individual needs and strengths relative to self
rather than others, and (d) integration of the results achieved with
the means by which they are achieved.

PMS has three parts: 1) NBC, 2) performance development,
and 3) evaluation and salary review. Each is carried out separately
and at different times, and performance appraisal is done in a three-
step process: (a) performance description questionnaire on which the
superior describes his observations of the subordinate's behavior, (b)
performance profile on which the superior analyzes the subordinate's
strengths and weaknesses, and (c) developmental interview during which
the superior attempts to help the subordinate see what changes are
needed so that together they may develop a plan for performance improve-
ment. Overall, this system has been well-accepted by both superior and
subordinate.104

Some support is given for reliability of the pair-comparison
technique when a rating criterion is to be used and it is known in ad-
vance that the group is homogeneous.105 Lawshe, et.a1., also concluded
that paired-comparison offered high interrater' reliability on the same

employee and also on successive ratings.106

 

104Michael Beer and Robert A. Ruh, "Employee Growth Through

Performance Management," Harvard Business Review, Vol. 54, July 1976,
pp. 59-66.
105 .
warren S. Blumenfeld and Sidney Q. Janus, "Interrater
Reliability of a Performance.Criterion for a very Homogeneous Group of
Managers," Psychological Reports, Vol. 35, December 1974, pp. 70-76.

106C. H. Lawshe; N. C. Kephart; and E. J. MCCormick, "The

Paired Comparison Technique For Rating Performance Of Industrial Em-
ployees," appearing in Thomas L. Whisler and Shirley F. Harper, Perfor-
mance Appraisal. (New York: Holt Rinehart and Winston, 1962),

pp. 209-15.

39
A number of studies have lent support to the concept of peer
appraisal. Kraut found peer rating to be a very fruitful way to predict
future success, even when peers knew each other for only a short time,

with reliabilities in the eighties (.80's) and validities in the thirties

to forties (.30's to .40's).107 Bayroff, et.a1., found peer rating by a

group more valid than rating by a single rater;108 Lewin and Zwany list

seven different studies supporting the position that peer rating has high

validity in the prediction of future performance.109 "In summary, peer

evaluations are valid tools for predicting future success and are super-

ior to all other measures available at the time of rating."110 There

are, however, factors which do apparently affect the validity of peer
ratings; these are summarized by Lewin and Zwany.

"1) Friendship does have a biasing effect upon the validity of peer
ratings in some situations.
2) Racial prejudices are apparently strong influences upon peer
ratings in most situation(s).
3) Length of acquaintanceship is important only as it provides in-
formation relevant to the rating criteria.
4) Interaction relevance is the crucial variable.
5) Face-to-face interaction is unnecessary so long as relevant be-
haviors are available to the raters.

 

107Allen I. Kraut, "Prediction Of Managerial Success By Peer

And Training Staff Ratings," Journal of Applied Psychology, Vol. 60,
February 1975, pp. 14-19.

108A. G. Bayroff; Helen R. Haggerty; and E. A. RundQUiSt,

"Validity of Ratings as Related to Rating Techniques and Conditions,"
appearing in Thomas L. Whisler and Shirley F. Harper, Performance
Appraisal. (New York: Holt Rinehart & Winston, 1962), pp. 112-125.

 

109Arie Y. Lewin and Abram Zwany, "Peer Nominations: A Model,

Literature Critique And A Paradigm For Research," Personnel Psychology,
Vol. 29, Autumn 1976, p. 423.

 

llOIbid., p. 430.

40
6) Within very broad limits of group homogeneity, peer evaluations
are stable across groups, and accurately predict future success
in situations where the group composition changes."111
Lee and Zwerman advocate an interesting expansion of the peer
rating approach to include a team approach. They suggest that each in-
dividual be evaluated by his superior, peers, and subordinates, and
that input from each of the three groups be weighted depending upon with
which group the ratee's interaction is most important.112
Other than the material on peer ratings, the literature does
not contain much regarding the person who does the rating. One study,
though, which involved the rating of one hundred (100) flight mechanics
by three levels of supervisors indicated that the level of raters closest
to the ratees was best able to rate them.113
All appraisal systems may be classified as either "absolute" or
"relative"; Duffy and webber discuss the limitations of each. Absolute
systems are those which rate an individual without reference to other
individuals but rather in regard to some absolute scale, while relative
systems rate an individual comparatively to other individuals in his

group. Though absolute systems often fail to discriminate and tend toward

"packing" at one end of the scale, relative systems also have serious

 

111Ibid., pp. 434-5.

112M. Blaine Lee and William L. Zwerman. "Designing a Mbtivating

and Team Building Employee Appraisal System," Personnel Journal, Vol. 55,
July 1976, pp. 354-57.

 

113Dean K; Whitla and John E. Tirrell, "The Validity of Ratings

of Several Levels of Supervisors," appearing in Thomas I» Whisler and
Shirley F. Harper, Performance Appraisal. (New York: Holt Rinehart &
Winston, 1962), pp. 132-35.

41
shortcomings in that they have excessive error across groups of small
size; furthermore it is difficult to make any kind of comparison between
a person within one group and a person in another.114 They conclude,
"As the ratee group size decreases, error increases, inversely as the
square root of group size."115 Also, "The solution to rating problems
with small groups is likely to lie in the direction of refinement of
absolute rating techniques."116

Behaviorally—anchored rating scales (BARS) has been the sub-
ject of a great deal of interest in performance appraisal in recent
years. Seemingly one of the most promising of appraisal methods, they
have not always lived up to that promise.

Briefly, the construction of a BARS is carried out in three
steps. First, groups of persons familiar with a given job (usually
supervisors and/or job holders) are asked to define behaviors or criti-
cal incidents which make up a job. Incidents which are not mentioned
by a given percentage (usually around sixty percent (60%) of the parti-
cipants are eliminated. Secondly, these are then placed in dimension
categories. Once dimension categories are established, they are clari-
fied to eliminate overlap. Third, a scale or continuum is developed

for each dimension listing Specific behaviors along the scale which

would be examples of poor, acceptable, good, or exceptional performance

 

114Kirt E. Duffy and Robert E. webber, "On 'Relative' Rating
Systems," Personnel Psychology, Vol. 27, Summer 1974, p. 307.‘

llsIbid., p. 308.

116Ibid., p. 311.

42

for that dimension.117

In developing BARS, Zedeck, et. al., found difficulty in deter-
mining examples of average or satisfactory performance on a poor-to-good
continuum. There was, for instance, disagreement between supervisors
and subordinates regarding the relative value of a given behavior.118

Borman and Dunnette, in comparing BARS with trait-oriented
performance ratings, found BARS somewhat superior in errors of leniency
and halo, and also in interrater agreement and the degree of differentia-
tion among raters. However, the magnitude of difference was small, never
over five percent (5%), and because BARS is costly in time and effort to
develop, they question whether it is worth the expense for so little
gain.119

Millard, by contrast, in a similar study found BARS to be
statistically superior to the graphic rating scale in avoiding errors of
halo, central tendency, and leniency.120

Finley, in comparing three scales --- two BARS (one behavior-

specific, one behavior-general) and a mixed-standard scale, found the

 

117H. John Bernardin; Mary B. LaShells; Patricia C' Smith; and

Kenneth M. Alvares, "Behavioral Expectation Scales: Effects of Develop-
mental Procedures and Formats," Journal of Applied Psychology, Vol. 61,
February 1976, pp. 75-9.

118Sheldon Zedeck; Nicholas Imparato; Moshe Krausz; Timothy =
Oleno, "Development of Behaviorally-Anchored Rating Scales as a Function
of Organizational Level," Journal of Applied Psychology, Vol. 59, April
1974, pp. 249-52.

119Walter C. Borman, "Behavior-Based vs. Trait-Oriented Perfor-
mance Ratings: An Empirical Study," Journal of Applied Psychology, Vol.
60, October 1975, pp. 561-5.

120CheedleWilliam Millard, The Development And Evaluation Of
Behavioral Criteria For Measuring The Performance of Non-Operational
Epployees. Doctoral Dissertation, university of Nebraska, 1974.

 

43

BARS superior in interrater reliability and in multi-trait, multi-rater
analysis of variance. He found the behavior-specific scale only slight-
ly superior to the behavior-general scale.121

Jordan, who used BARS development techniques in a survey of
administration and faculty regarding the dimensions of a department
chairman's job, concluded that, (a) BARS methodology is a potentially
useful tool for the development of criteria and measures for the assess-
ment of an administrator's performance in higher education, (b) that the
scales developed represent useful measures of a chairman's performance,
and (c) that ratings of a chairman's performance are influenced by a com-
plex set of factors and interactions which need to be considered when
rating instruments are used to assess that performance.122

Perhaps the most definitive literature survey on BARS is that
done by Schwab, et.al. They state that while BARS is used less frequent-
ly than other types of measures, it has been recommended by numerous in-
vestigators because the evaluator has to make fewer inferences about the
employee, thereby serving more as an observer and less as a judge.

They see the develOpment of a BARS as a five-step process:

1) Critical incidents - A group of persons with knowledge of the
job are asked to describe specific illustrations of effective
and ineffective performance.

2) Performance dimensions - The instrument developers then cluster
the incidents into a smaller set of performance dimensions which

they typically define (usually five (5) to ten (10).
3) Retranslation - Another group of job-knowledgeable participants

 

121David mgrle Finley, The Effects Of Scale Continuity And

Behavioral Anchor Specificity Upon The Psychometric Properties Of Per-

formance Rating Scales. Doctoral Dissertation, University of Houston,
1975.

122Jackie.B..Jordan, The Development And Evaluation Of Behavior-
ally Anchored Rating Scales For The Assessment Of Department Chairman Per-
formance. Doctoral Dissertation, University of Kentucky, 1976.

44

is instructed to retranslate (or reallocate) each critical in-
cident to the dimension it best describes. An incident is re-
tained typically if fifty (50) to eighty-eight (88) percent of
the group assigns it to the same dimension as group number one.

4) Scaling incidents - The group is also asked to rate (usually
on a seven (7) - nine (9) point scale) the behavior described
as to how effectively or ineffectively it represents performance
on a given dimension. A standard deviation criterion of agree-
ment for scale placement of an item (usually one point fifty
(1.50) or less on a seven (7) point scale) is set to determine
which incidents will be retained.

5) Final instrument - A subset of incidents (usually six (6) or
seven (7) Per dimension) that meet both retranslation and stan-
dard deviation criteria are used as behavior anchors for each
performance dimension. The final BARS consists of a series of
vertical scales for each dimension, anchored by specific inci-
dents, each located on the scale based on its rating established
in Step 4.123

 

BARS have a number of hypothesized advantages:

1) The scales are highly job-oriented.

2) There are high levels of user participation in the development
of the instrument.

3) They are developed by people who know the job better than any-
one else.

4) Incidents are highly specific and nonambiguous.

5) They should reduce errors of leniency and central tendency.

6) Participation in develOpment means people are more likely to
accept the results.

7) Specific incidents can be used as a development or training
tool.

Schwab, et.a1., conclude that the research to date (1975) has
not conclusively demonstrated these advantages, although it should be
noted that research since then, some of which has been noted earlier in
this chapter, has tended to be supportive, though not conclusive.124

Keeley believes that there are circumstances which limit the

usefulness of BARS, however. He believes that they can be developed

 

123Donald P. Schwab; Herbert G. Heneman III; and Thomas A-

DeCotiis, "Behaviorally Anchored Rating Scales: A Review Of The Litera-
ture," Personnel Psychology, vol. 28, Winter 1975, pp. 550-1.

124Ibid., pp. 551-3.

45

only where tasks are clear and there is a high degree of consensus re-
garding what constitutes good and-bad performance.12S

Goodale and Burke believe that four criteria should be applied
in determining what dimensions should be included in a BARS. These are:
(a) a dimension must be broad enough to be worth evaluating, but not so
broad as to be meaningless, (b) it has to be observable, (c) it must be
applicable to all individuals being evaluated, and (d) has to be relevant
to a large number of jobs.126

The graphic rating scale has been criticized in some studies
when compared to BARS, but it has been criticized also on logical and
practical grounds as early as 1945 by Ryan. "This article has supported
the thesis that the graphic rating scale with numerical scoring has
serious logical difficulties and practical inadequacies. These flaws
are so serious that they more than offset any advantages which are
claimed for these ratingtechniques."127

Two additional performance appraisal techniques are of interest.
One, presented by Salton, makes use of modern computer technology to

tailor performance appraisal to fit each job. He suggests programming a

computer with a variety of appraisal items from which could be selected

 

125Michael Keeley, "Subjective Performance Evaluation And Per-

son-Role Conflict Under Conditions Of Uncertainty," Academy of Manage-
ment Journal, Vol. 20, No. 2, June 1977, p. 312.

 

126James G. Goodale and Ronald J. Burke, "Behaviorally-Based

Rating Scales Need Not Be Job Specific," Journal of Applied Psychology,
Vol. 60, June 1975, pp. 389-91.

 

127Thomas A. Ryan. "Merit Rating Criticized," appearing in

Thomas L. Whisler and Shirley F. Harper, Performance Appraisal. (New
York: Holt Rinehart & Winston, 1962), p. 169.

 

46

by the evaluator those appropriate to a given job. These items could
also be weighted as appropriate. The evaluations would then be placed
in the computer for future recall and reference. He believes that this
system offers the following advantages: (a) it will facilitate perfor-
mance improvement because each system is applicable to a specific job,
(b) it will better meet the evaluator's needs, (c) it is easily main-
tained and updated, (d) it is easily administrated (all evaluations are
in the computer), and (e) it is simply and easily introduced.128

The other is offered by White and Barnes, although it appears
to be more in the nature of a supplement to, rather than a replacement
for, other forms of appraisal. They recommend the formation of groups
of subordinates or superiors, or combinations of both, to discuss solving
organizational problems and to examine the interdependency of each and
all levels within the organization, and also to look at the quality of
that interdependency. In this way, they feel, each person will see how
and where he fits in the organizational structure, and performance
appraisals will stress that fact.129

As can be seen, there are many methods of performance appraisal,
each with its strengths and weaknesses. Selection of a method should be

done carefully, taking into consideration its purpose, its strengths and

weaknesses, and the needs and resources of the organization.

 

128Salton, op.cit., p. 58.

129White and Barnes, op.cit., pp. 108-9.

47

Other Factors That Affect
Performance Appraisal

 

 

Some research has considered the effects of race and sex in an
appraisal situation. Bass and Turner, in reviewing evaluations of black
and white bank tellers (raters were primarily white), found no signifi-
cant differences when age and job tenure were held constant. The re-
searchers did find that on salary recommendations, supervisors tended to
consider objective factors more with blacks than whites, possibly because
they may be sensitive to racial tensions. They also quote a 1969 study
by Flaugher and Norris which showed similar results, and a 1966 study
(Rotter and Rotter) which indicated a tendency to rate blacks higher when
performance was poor, but showed no difference between ratings of blacks
and whites when performance was good. Bass and Turner believe that the
ratings of blacks by whites often fails to consider less tangible factors
such as motivation and attitude which are considered for the whites; in
such cases, ratings are often not comparable. They recommend that the
organization clearly define what constitutes success on the job and then
make sure measurement criteria are the same for all employees, concen-
trating primarily on elements where equivalence across groups is known,
generally objective data.130

Bigoness, using sixty (60) white male undergraduate students,
had them view a videotape of a white male and female, and a black male
and female, stacking cans. Each subject wore identical clothes and used

the same gestures. The subjects were either high performance (stacked

 

lBOAlan R. Bass and John H. Turner, "Ethnic Group Differences

in Relationships Among Criteria of Job Performance," Journal of Applied
Psychology, Vol. 57, April 1973, pp. 101-9.

 

48

forty-eight (48) cans) or low performance (stacked twenty-four (24)
cans). Bigoness found that low performance males and females were
rated identically, but that high performance females were rated signi-
ficantly higher than males. He also found that low performance blacks
were rated higher than low performance whites; however there was no
significant rating difference for high performance between blacks and
whites, thus supporting the Rotter study mentioned previously.131

Feild and Holley, who examined appraisal results in an organi-
zation where discussion of the results with the employee was optional,
found that age and sex had a negligible effect on whether the results
were discussed, though race had a definite effect. (All raters were
white). For blacks, raters tended to discuss high ratings but to avoid
discussing low ratings. For whites, the pattern was the reverse.132

As was indicated earlier in the chapter, subordinates frequent-
ly view the purpose of the performance appraisal differently than do
supervisors. LaBean, in a survey done at Dow Chemical, found that nearly
fifty percent (50%) of the subordinates who responded felt that insuffi-
cient time in the appraisal was devoted to such areas as career oppor-

tunities, how the supervisor could help the subordinate, how much the

subordinate had improved in performance in the past year, and how the

 

131WilliamJ. Bigoness, "Effect of Applicant's Sex, Race and
Performance on Employer's Ratings: Some Additional Findings," Journal
of Applied Psychology, vol. 61, February 1976, pp. 80-4.

132Hubert S. Feild and William H. Holley, "Subordinates'

Characteristics, Supervisors' Ratings, and Decisions to Discuss Appraisal
Results," Academy of Management Journal, Vol. 20, No. 2, June 1977,
pp. 315-21.

49

subordinate could continue to improve.133 a

Summegy

The literature on performance appraisal has been reviewed in
this chapter looking at Opinions on the need for it, problems with it,
and some recommendations to deal with or alleviate these problems.
Lastly, a representative variety of methods Of performance appraisal
have been examined, their strengths and weaknesses, and some factors
which affect performance appraisal.

While Opinion is divided on the merits of performance appraisal,
most authorities, superiors, and even subordinates feel that performance
appraisal is necessary, in spite of the fact that their respective views
of its purpose may be divergent and even though performance appraisal
does not always accomplish its desired purpose(s).

Some of the problems with performance appraisal result from the
fact that the evaluation Of a perSon's job performance is an emotional
issue, especially if the evaluation is critical in nature. Often, as was
indicated, criticism results in a decline in performance, the Opposite
effect Of that intended. Performance appraisal appears especially weak
in dealing with an employee who is performing below standard.

Another area of difficulty revolves around the evaluator,
especially as personal biases and inconsistencies bear on the evaluation.
Frequently, too, the evaluator is poorly prepared and/or trained for this

role.

Yet other problems result from confusion of purpose or the

 

133James H. LaBean, Employee Perceived values Of JOb Perfor-

mance Reviews. Masters Thesis, Central Michigan university, 1975.

50
attempts by management to use an appraisal system inappropriately or to
serve tOO many purposes.

In spite of these problems, performance appraisal continues to
be used by most organizations and a great deal Of time and effort has
been spent in attempting to make it work more productively.

Some suggestions that have proven fruitful are to keep the
appraisal system simple, clearly defined, and participative in nature.

It is important that the superior and subordinate view performance
appraisal as an opportunity for growth and not an authoritarian critique.
Some additional approaches that show promise are peer appraisal and
training for evaluators.

There is a great variety Of methods for performance appraisal,
each with its relative merits and demerits. The consensus seems to be
that effective appraisals should lean most heavily upon objective
measures, but that subjective measures cannot be ignored.

While it would appear that rating scales of various types are
the most frequently used method, the essay approach is still fairly wide-
ly used and MBO is making some inroads, although most authorities do not
feel that it can be used to supplant performance appraisal.

One of the fastest-growing and most promising methods developed
in recent years is the behaviorally-anchored rating scale. While it is
difficult and time-consuming to construct and has had its share of grow-
ing pains, it shows considerable promise in cutting down on the most
common rating errors.

A certain amount Of research has dealt with the question of how
such factors as age, sex and race affect performance appraisal. While

age does not appear to be much of an influencing factor, both sex and

51

race do seem to have an influence.

CHAPTER III

RESEARCH METHODOLOGY AND
DESIGN OF THE STUDY

ose

The purpose Of this study was to substantiate that differ-
ences existed in the perceptions of performance appraisal by superior
and subordinate in regard to purpose, need, and method.

Selection Of Research Method

 

The research method selected was a survey questionnaire design-
ed by the researcher (Appendix, Exhibit B). The questionnaire was deve-
lOped through consultation with an expert in questionnaire design along
with four corporate personnel specialists, who examined the instrument
for face validity. Reliability, clarity, and freedom from ambiguity
were tested by administering the questionnaire to a test sample (N = 20)
from a small industrial concern. The design contained in the appendix
is the final one developed and was used in the study.

The questionnaire was distributed to all full-time home Office
employees (N = 565) Of an internationally-known corporation located in
the Midwestern united States. Distribution was by company mail. TO
each questionnaire was attached a cover letter (Appendix, Exhibit A),
the content Of which was approved by an expert in questionnaire design
and the company management, and a stamped, self-addressed envelope. The

data was collected during the month Of November, 1977.

52

53

Testing Of Hypotheses And
Instrument Design

 

 

Questions 1 through 8 solicit information regarding the re-
spondent's classification Of supervisor/subordinate, salary exempt/non-
exempt, years on job, age, education, sex, and ethnicity.

Questions on the survey instrument are keyed to hypotheses to
be tested in the following manner:

Hypothesis 1:

The need for performance appraisal will be seen as more important
by supervisors than subordinates.

9. Some type of performance appraisal should be carried out for
all employees.

strongly agree disagree
agree strongly disagree

Question #9 tests this hypothesis by showing the relative im-
portance Of appraisal to the respondent. Those who see it as more im-
portant will tend tO agree with this statement; those who see it as less
important will tend to disagree with this statement.
10. If you agree, how Often should this appraisal be carried out?

twice a year every two years
annually other (specify)

 

 

Question #10 tests this hypothesis by checking the desired
frequency of performance appraisal. Those who see performance appraisal
as more important will feel that it should be carried out more frequently
than those who see it as being of lesser importance.

Hypothesis 2:

The preferred method for carrying out a performance appraisal will
be perceived differently by supervisors and subordinates.

A. Supervisors will prefer a method which maximizes their
input.

54

B. Subordinates will prefer a method which maximizes their

input.

C. Supervisors will prefer a method which gives more weight
to subjective criteria.

'D. Subordinates will prefer a method which gives more weight
to Objective criteria.

11. A performance appraisal should be done by: (check one)

the immediate supervisor only

the immediate supervisor and the person supervised (self-
evaluation)

the department head only

the department head and the person supervised (self-
evaluation)

the department head and the immediate supervisor

the person being evaluated only (self-evaluation)

a person from outside the company (example, consultant)
the department head, immediate supervisor and person
supervised (self-evaluation)

other (specify)

 

 

 

 

 

 

 

Question #11 tests Hypothesis 2, A and B. The responses to
this question are graduated from that which allows maximum input from
the subordinate - "the person being evaluated only (self-evaluation)" -
to "the department head only", which most minimizes input from the subor-
dinate, since it relies upon input from the person most distant from the
subordinate. The response, "a person from outside the company (example,
consultant)", is seen as essentially neutral, since it does not rely
primarily upon input from either superior or subordinate.

14. Performance appraisal can be carried out using Objective and/or
subjective criteria. Examples of Objective criteria might be:
quantity Of work done, objectives met, etc. Examples Of sub-

jective criteria might be: employee attitudes, personality
traits, quality Of work done, etc.

 

 

 

In your opinion, performance appraisal should be based on:
(check one)

55
Objective criteria only
subjective criteria only
primarily Objective, but include sOme subjective criteria
primarily subjective, but include some Objective criteria
equally based on both subjective and Objective criteria
Question #14 tests Hypothesis 2, C and D, and responses are
graduated from the utilization Of completely Objective, to completely

subjective, criteria.

Hypothesis 3:

 

The pppposes Of performance appraisal will be perceived differently
by supervisors and subordinates.

A. Supervisors will agree that salary increases should be tied
directly to performance appraisal.

B. Subordinates will feel that salary increases should not be
tied directly to performance appraisal.

C. There will be a significant difference in the rank order Of
the purposes Of performance appraisal for supervisors and
subordinates.

12. An individual's salary increase should be tied directly to his/
her performance appraisal.

strongly agree disagree
agree strongly disagree

13. Listed below are six (6) possible purposes for or uses Of per-
formance appraisal. Please rank them in order Of importance as
you see them with 1 being highest and 6 lowest. ‘

provide feedback to employee on job performance
assist in employee's professional growth
determine salary merit increases
determine promotion within organization
rovide an ongoing record Of employee's growth and develop-
ment -
improve communications between supervisor and employee

Question #12 will test Hypothesis 3, A and B, and Question #13

will test Hypothesis 3, C.

56

Selection Of Sample

 

Several criteria were considered important in the sample
selected to give validity to this study. The sample should be composed
of: (a) respondents who were familiar with performance appraisal, and
(b) a mix Of supervisors and subordinates.

Although not directly related to the main investigation of

134, 135, 136
an

this study, other researchers had indicated that race d

sex137 affected performance appraisal. There does not appear to be any
significant research on the effect Of educational level and age upon
performance appraisal, other than an indication by Feild and Holley that
age did not seem to affect whether or not performance appraisals were
discussed with the employee.138 It was decided, as a result Of a review
of these studies, to include the factors Of race, sex, age, and educa-
tional level as adjunctive parts to this study, and, therefore, an attempt
was made to find a sample which had a good mix Of these factors, as well
as meeting the major criteria described in (a) and (b) above.

In addition, biographical material regarding the composition
Of the sample had to be available to the researcher; this was: (a) sample
size, (b) sample breakdown regarding the number of supervisors and subor-

dinates, and (c) a sample breakdown for sex, race, age, and educational

 

134Bass and Turner, op.cit., pp. 101-9.

135Bigoness, op.cit., pp. 80-4.

136Feild and Holley, op.cit., pp. 315-21.

137Bigoness, op.cit., pp. 80-4.

138Feild and Holley, op.cit., pp. 315-21.

57
level. Finally, Of course, the sample had to be accessible to the re-
searcher and be willing to participate in the study.

Several survey populations were considered and rejected, either
because Of not meeting the majority Of the criteria or because Of failing
to meet some of the more critical criteria. The survey population
settled upon is the full-time home Office employees of an internationally-
known corporation located in the Midwestern united States. 'The total home
Office employee N = 565. The sample has two fairly notable limitations:
(a) only five percent (5%) Of the home Office employees are members Of a
racial minority (four percent (4%) black), and fifty-four percent (54%)
have a college degree. An additional twenty-six percent (26%) have some
college and only twenty percent (20%) of the sample have a high-school
education or less.

Review System

 

Formal written performance appraisal has been carried out in
this company since 1965, and the present fOrmat has been in use fOr
approximately ten years. There is some variation in format for profes-
sional (generally salary-exempt) employees and Office and technical
(usually salary non-exempt) employees. The company uses the term "review"
rather than "appraisal".

The performance review form used for professional employees is
a modified MBO approach with an essay format. It includes a percentile
ranking for the individual within his group. If his responsibilities in-
clude interface with two or more managers, there may be a dual review of
the employee. The employee does not, himself, fill out a self-evaluation
form but there is some space for his comments on the review form. The

employee receives a copy of the completed review. Instructions for

58
filling out the rating form are included on the form.

The perfOrmance review form for Office and technical employees
uses an essay format. It includes a sheet for discussion Of the employ-
ee's perfOrmance on which both employee and supervisor write comments
dealing with specific areas. This is not part Of the final summary re-
view form but there is some space for employee remarks on the summary
form. The employee does not receive a percentile ranking; however, he
does receive a copy Of the completed review form.

Sample Description

 

A. Educational Level

Salaried exempt’: Some high school 1
High school - 21

Some college 49

BS/BA 170

MS/MA 105

Ph.D. 19

Total 365

Salaried non-exempt": Some high school 18
High school 74

Some college 96

BS/BA ' 12

Total . 200

Grand Total Employees 565

 

‘ Salaried-exempt - Those employees who, under the Fair Labor
Standards Act are not eligible for overtime pay for overtime work.
Generally they are college graduates who hold a professional or super-
visory position or are involved primarily or substantially in making
policy decisions or using their professional training.

* Salaried non-exempt - Those employees who, under the Fair
Labor Standards Act are eligible for overtime pay. They are usually
non-professional, non-college-educated employees.

 

59

B. Sex and Race

Spanish-
Speaking American
Caucasian Black American Indian Oriental
Male 360 9 4 - 2
Female 175 ll 2 2 -
C. Age
Exempt" NOn-exempt’ Total
under 20 - 5 5
20-29 48 83 131
30-39 159 49 208
40-49 95 38 133
50-59 50 23 73
60 + 13 2 15
Total 365 200 565

The survey population is ninety-five percent (95%) white,
sixty-six percent (66%) male, fifty-four percent (54%) college degreed,
and has the following age breakdown: twenty-four percent (24%) under
29, thirty-seven.percent (37%) between 30-39, twenty-three point five
percent (23.5%) between 40-49, thirteen percent (13%) between 50-59,
and two point five percent (2.5%) over 60. The mean age Of the popula-
tion is thirty-nine point eight (39.8) for exempt and thirty-four point
four (34.4) for non-exempt employees or thirty seven point nine (37.9)
for all employees. An exact breakdown Of the number Of supervisors was
not available, but the company personnel department feels that a reliable
estimate would be two hundred and twenty-five (225). A broad range Of
occupational strata is represented, excluding heavy industrial types of

occupations.

 

' See previous page.

60
Analysis

The statistical techniques applied to the data are those pro-
grammed by the Statistical Package for the Social Sciences (SPSS) and
were completed by the Michigan State Computer Services.

The chi square test allows the investigator to determine
whether the Obtained frequencies in a set Of categories differ signifi-
cantly from the expected frequencies. This test was applied to the
reSponses of the two groups (SUpervisor and subordinate) on Questions
9, 10, ll, 12 and 14. The SPSS package also allows for computation of
the degree of significance along with the raw chi square. This was com-
pleted on each item.

The Spearman coefficient Of correlation allows for the compari-
son Of the ranking of a number of items by two groups Of respondents.
The coefficient of correlation measures the degree to which the rankings
Of a test item by each group correlates with the other. This test was
applied to the responses of the supervisor and subordinate groups to
Question 13.

Cross tabulations were carried out comparing the independent
variables Of age, sex, race, and educational level, with the responses
to the questionnaire (dependent variables) to determine whether or not
these independent variables had a significant effect upon responses.
Since these factors were not part Of the hypotheses selected fOr this
study, the results Of the cross-tabulations were covered in the Bie-
cussion section of Chapter IV.

Assumptions

 

1. The survey instrument design adequately tests the hypotheses.

2. The survey instrument is free from bias, clear, and non-ambiguous

61
for all respondents.
3. The sample selection criteria are valid.
4. The cover letter was read and understood by the respondents.
5. The respondents saw this as an independent study, not carried
out by or for the company for which they worked.
6. The respondents answered the questions honestly.
7. The survey questionnaire method is the best method for testing
the hypotheses.
8. The survey population can reasonably be expected to adequately
test the hypotheses.
9. The methods of analysis selected are appropriate to the design
Of the study.
10. The respondents understood the terminology used in the survey
instrument.
SummaEy
This study, using the survey questionnaire method, tested the
hypotheses by administering the questionnaire to a survey population
consisting Of the home Office employees (N = 565) of an internationally-
known corporation located in the Midwest during the month of November,
1977.
Methods Of analysis consisted of chi square, Spearman co-
efficient Of correlatiOn, and a cross-tabulation Of the results on inde-

pendent variables in combination with dependent variables.

CHAPTER IV

ANALYSIS OF THE DATA

Response Rate

 

Consultants in the Michigan State University Office of Educa-
tional Research stated that a response rate of fifty percent (50%) to
sixty percent (60%) would be acceptable for this study. The minimum
recommended level Of significance for any item was point zero five
(.05). These two recommendations were followed in this study.

The questionnaire was distributed tO all company personnel on
November 15, 1977. The cut-Off date selected was December 6, 1977,
three weeks later. As Of the cut-Off date, a total Of three hundred
eighty-eight (388) responses or sixty-nine percent (69%) Of the total
N Of 565 had been received.

Of the two major sub-groups in the population, superiors and
subordinates, one hundred eighty-six (186) superiors, or eighty-three
percent (83%), and two hundred and two (202) subordinates, or fifty-nine
percent (59%), responded. Three additional responses had to be rejected
because the respondents did not specify whether they belonged to the
category Of superior or subordinate.

Analysis Of Results

 

Hypothesis 1:

 

The need for performance appraisal will be seen as more important
by supervisor than subordinate.

62

63
Question #9 on the survey instrument tested this hypothesis by
determining to what degree the respondents felt performance appraisal
was necessary. Ninety-nine point five percent (99.5%) Of all respondents
agreed that perfOrmance appraisal should be carried out on all employees.
However, as can be seen by the following table, superiors agreed more

strongly with this statement than did subordinates.

TABLE 4.1

RESPONSES TO QUESTION #9

 

Agree Strongly Agree
Su erior N 35 148
p % 18.9 80.0
. N 63 139
Subordinate % 31.2 68.8

 

With a raw chi square of nine point fifty-five (9.55), the
degree Of significance was point zero-zero-eight-four (.0084). The re-
sponse to Question #9 supports Hypothesis 1.

Question #10 tested Hypothesis 1 by determining how Often the
respondents felt performance appraisal should be carried out, with great-
er frequency being equated with greater importance. Ninety-six point one
percent (96.1%) Of all respondents felt that performance appraisal should
be carried out on either an annual or a bi-annual basis. The following
chart indicates, however, that superiors gave greater importance to bi-

annual evaluations than did subordinates.

64

TABLE 4.2

RESPONSES TO QUESTION #10

 

 

 

Bl-annual Annual
S erior N 115 61
up % 62. 2 33.0
N 106 90
Subordinate % 52.5 44.6
Column percentage 57.1 39.0

 

With a raw chi square of nine point thirteen (9.13) for all

responses tO Question #10, the degree Of significance is point zero-two-

seven-six (.0276). The response to Question #10 supports the hypothesis.

Hypothesis 1 is accepted.

Hypothesis 2:

 

The preferred method for carrying out a performance appraisal will
be perceived differently by supervisors than subordinates.

A.

B.

C.

D.

Supervisors will prefer a method which maximizes their
input 0

Subordinates will prefer a method which maximizes their

input.

Supervisors will prefer a method which gives more weight
to subjective criteria.

Subordinates will prefer a method which gives more weight
to Objective criteria.

Question #11 tested Hypothesis 2 and also sub-hypotheses A and

B by determining whom the respondents felt Should carry out the perfor-

mance appraisal.

65

 

HCOEO>HO>CH MOHumosm oumzoa DCOEO>HO>EH wuocHouonsm phosoe

 

 

 

A v
m. o.m m.OH m. b.0m o.mm m. o.m x oumcHouOQOm
H v mm H mm mOH H w z
m. m.m m.eH m. N.Hm m.mm o H.@ a MOHMOOSm
H HH hm H pm me o mH 2
Eco 3:0 Eco .6an 3% a 3mm e 38 umfio
comm uOmH>uoosm uOmH> upso uOmH>uwmom uOmH> MHom
.pooo a prom .uaoa unwozm .omom .uemo iuoesm

 

 

 

I'll-1“ [Inn's-ll II 'lllllllllill

HH# ZOHBmmDO OB mmmzommmm

m.¢ mqm<e

66

While the trend for all respondents was toward a system in
which the ratee participated, the trend was stronger in this direction
for subordinates than for superiors. Eighty-three percent (83%) Of the
subordinates favored a system in which the ratee was directly involved,
while seventy point four percent (70.4%) Of the superiors favored such
a system. With a raw chi square Of thirteen point eighty-eight (13.88),
the degree Of Significance is point zero-five-three-three (.0533). The
response to Question #11 supports Hypothesis 2 and also sub-hypotheses
A and B.

Question #14 tested Hypothesis 2 and sub-hypotheses C and D by

determining the type of performance appraisal criteria preferred by the

 

 

respondents.
TABLE 4.4
RESPONSES TO QUESTION #14
Equally
Objective Primarily Objective/ Primarily Subjective
Only Objective Subjective Subjective Only
Su erior N 3 97 68 13 l
p % 1.6 53.3 37.4 7.1 .5
. N 8 58 107 29 0
smrdlnate % 4.0 28.7 53.0 14.4 0
Column percentage 40.4 45.6 10.9

 

Although there is a marked difference in the criteria preferred
by the two groups Of respondents, it is not in the predicted direction.
Superiors prefer Objective criteria, while subordinates prefer most

strongly a method which gives equal weight to both Objective and

67
subjective criteria. With a raw chi square Of twenty-six point ninety
(26.90), the degree Of significance is point zero-zerO-zerO-zero (.0000).

The response to Question #14 supperts Hypothesis 2, but does not support

 

sub-hypotheses C and D.

 

Hypothesis 2 is accepted. Sub-hypotheses 2, A and B are

 

 

accepted. Sub-hypotheses 2, C and D are rejected.

 

 

Hypothesis 3:

 

The ppppose Of performance appraisal will be perceived differently
by supervisors and subordinates.

A. 'Supervisors will agree that salary increases should be
tied directly to performance appraisal.

B. Subordinates will feel that salary increases should not
be tied directly to perfOrmance appraisal.

C. There will be a significant difference in the rank order
of the purposes Of performance appraisal fOr supervisors
and subordinates.

Question #12 tested Hypothesis 3 and sub-hypotheses A and B by
determining whether or not the respondents felt that salary increases
should be tied directly to performance appraisal. A total Of eighty-
eight point six percent (88.6%) of all respondents agree that salary
increases should be tied directly to performance appraisal. Although

this trend was stronger for superiors, it was nevertheless true for sub-

ordinates as well.

68
TABLE 4.5

RESPONSES TO QUESTION #12

 

 

 

 

 

Strongly Strongly
Disagree Disagree Agree Agree
S . r N 5 9 97 74
”perlo % 2.7 4.9 52.4 40.0
. N 4 26 109 62
S“b°rdlnate % 2.0 12.9 54.2 30.8
Column percentage 2.3 9.1 53.4 35.2

 

With a raw chi square Of nine point forty-eight (9.48), the

degree Of significance was point zero-two-three-six (.0236). The response

 

to Question #12 does not support Hypothesis 3. It does support sub-

 

Hypothesis 3 A, but does not support sub-Hypothesis 3 B.

Question #13 tested Hypothesis 3 and also sub-hypothesis 3 C
by determining the respective ranking by supervisors and subordinates Of
six items regarding the purposes of performance appraisal. The mean rank-
ing for each item was computed for the superior and subordinate groups,
and then a Spearman correlation coefficient was run comparing the rank-
ings. A positive correlation of rankings was fOund (pOint eight-eight-
five-seven (.8857) with a significance level of point zero-one-zero (.010).

Rankings Of purpose were as follows:

69

TABLE 4.6

RESPONSES TO QUESTION #13

 

 

Superior _ Subordinate

 

1) TO provide feedback to employee 1) TO provide feedback to employee
2) TO aid employee's professional 2) TO determine salary merit

growth increases

3) To determine salary merit 3) To aid employee's professional
increases ‘ growth

4) TO improve supervisor-employee 4) TO determine promotion
communication 5) TO improve supervisor-employee

5) To determine promotion communication

6) TO provide an ongoing record 6) To provide an ongoing record

 

As can be seen from the table above, the ranking is very similar
for both groups and differences are never more than one rank apart. £233
responses to Question #13 support neither Hypothesis 3 nor sub-hypothesis
§_S§

Hypothesis 3 is rejected. Sub-hypothesis 3 A is accepted; sub-
hypotheses 3 B and C are rejected. I

Discussion

 

The response to Question #9 supports the findings reported in
the literature that both superiors and subordinates feel that performance
appraisal is necessary. In fact, one hundred percent (100%) Of the sub-
ordinates who responded felt this way.

The response tO Question #10 indicates that a clear majority
Of respondents (fifty-seven point one percent (57.1%), favored bi-annual
performance appraisals over any other type. This preference is more
marked for the superior group, who showed sixty-two point two percent

(62.2%) in favor of bi-annual appraisals.

70
It should also be noted that there appears to be a desire for
the bi-annual over the annual review as the level Of education Of the
respondent rises (see table below), and the converse as the educational

level goes down.‘

TABLE 4.7

RESPONSES TO QUESTION #10

 

Education Bi-annual (%) Annual (%)
High School 53.7 43.9
Some College . 56.8 40.7
Bachelor's 54.3 40.2
Master's 58.1 39.5
Ph.D. 92.3 0

 

Question #11 responses revealed a very marked preference on the
part Of both superiors and subordinates for a method in which the ratee
was directly involved by evaluating himself, with a total Of seventy-
seven point one percent (77.1%) favoring this method. This is interest-
ing since in the company in which the survey was done, only salary non-
exempt employees make self-appraisals and even then, these are not part
of the final form. These responses, however, do coincide with the recom-
mendations in the literature which advise that self-appraisal should be a
part Of the performance appraisal process.

A considerably greater percentage Of women (fifty-five point

 

‘ It should be noted here that the discussion Of independent
variables, other than superior and subordinate, is limited to those
questions where it was found that these variables Significantly affected
the response.

71
three percent (55.3%) than men (forty-one point seven percent (41.7%)
favor appraisal by the supervisor and the ratee, while a greater per-
centage Of men (fourteen point eight percent (14.8%) than women (seven
point three percent (7.3%) favored appraisal by the supervisor alone.

The responses to Question #12 reveal a surprisingly strong
feeling that salary increases should be tied directly to performance
appraisal. However, this feeling is considerably stronger among superiors
than subordinates, with ninety-two point four percent (92.4%) of the
superiors in agreement as Opposed to eighty-five percent (85%) Of the
subordinates. A considerably larger percentage Of women (fifteen point
five percent (15.5%) than men (nine point five percent (9.5%) disagree
with the concept Of tying salary increases directly to performance
appraisal.

Although the ranking of purposes of perfOrmance appraisal is
consistent between superior and subordinate groups with very little differ-
ence, there is a marked difference between the sexes in the ranking Of

these items as a look at the following table shows.

72
TABLE 4.8

RESPONSES TO QUESTION #12

 

-——————————-——-— —_—_—-—_-———-—--———-——————-‘_—--——_—
—-.—_——————_——-— — —---- — u.- — -—-—-

 

Male Female

 

1) TO provide feedback to employee 1) To provide feedback tO employ-

2) TO aid employee's professional ee

growth 2) TO improve supervisor-employee
3) To improve supervisor-employee communication

communication 3) TO determine salary merit
4) TO determine salary merit increases

increases 4) TO aid employee's professional
5) TO determine promotion growth
6) TO provide an ongoing record 5) To determine promotion

6) TO provide an ongoing record

 

The response Of the subordinate group to Question #14 was sig-
nificant, with only thirty-two point seven percent (32.7%) feeling that
Objective or primarily objective criteria Should be given greatest weight
in performance appraisal and fifty-three percent (53%) favoring an even
mix of Obj ective and subjective criteria.

There is a very definite difference between college-graduate
and non-college-graduate respondents on this question. High-school
graduates (sixty-one percent (61%) and employees with some college (fifty-
Seven point three percent (57.3%) favor an equal mix Of Objective and sub-
jective criteria; however, bachelor's degree, master's degree, and Ph.D.
level respondents favor primarily Obj ective criteria at percentages Of
fifty-two point four (52.4), fifty-five point three (55.3), and sixty-one
point five (61.5), respectively.

Additionally, males favor the use Of primarily Objective criteria
much more (forty-six point five percent (46.5%) than do females (twenty-

seven point six percent (27.6%).

73
It should be noted that the response rate varied somewhat

according to educational level (see table below).

TABLE 4.9

RESPONSES TO QUESTION #14

 

 

Some high school 5% Bachelor's degree 70%
High school 43% Master's degree 83%
Some college 81% Ph.D. 68%

 

The response rate for minorities was low, with thirty percent
(30%) Of blacks and seventeen percent (17%) Of Spanish-speaking Americans
responding. This, coupled with their small percentage Of the total sam-
ple, prevents any inferences being drawn along ethnic lines.

m

Sixty-eight percent (68%) of the sample responded. Of these,
the two major sub-groups, supervisors and subordinates, responded at the
rates Of eighty-three (83) and fifty-nine (59) percent respectively.

Hypothesis 1:

 

The need for performance appraisal will be seen as more important by
supervisors than subordinates.

Hypothesis 1 was accepted. This hypothesis was supported by
responses to Question #9, (significance level, point zero-zerO-eight-four
(.0084) and Question #10 (Significance level, point zero-two-seven-six
(.0276).

Hypothesis 2:

 

The preferred method for carrying out a performance appraisal will
be perceived differently by supervisors than by subordinates.

Hypothesis 2 was accepted. This hypothesis was supported by

74
the responses to Question #11 (significance level, point zero-five-three-
three (.0533) and Question #14 (significance level, point zero-zerO-zero-
zero (.0000).

Sub-hypothesis 2, A:

 

Supervisors will prefer a method which maximizes their input.

Sub-hypothesis 2, B:

 

Subordinates will prefer a method which maximizes their input.
These hypotheses were supported by the responses to Question
#11 (significance level, point zero-five-three-three (.0533), and were
accepted.

Sub-hypothesis 2, C:

 

Supervisors will prefer a method which gives more weight to subjec-
tive criteria.

Sub-hypothesis 2, D:

 

Subordinates will prefer a method which gives more weight to Objec-
tive criteria.

These hypotheses were not supported by the responses to Question
#14 (significance level, point zero-zerO-zero-zero (.0000); consequently
they were rejected.

Hypothesis 3:

 

The ppppose Of performance appraisal will be perceived differently by
supervisors and subordinates.

This hypothesis was supported by neither the responses to Ques-
tion #12 (significance level, point zero-two-three-six (.0236) nor Ques-
tion #13, positive correlation point eight-eight-five-seven (.8857) (sig-
nificance level, point zero-one-zero (.010), and therefore was rejected.

Sub-hypothesis 3, A:

 

Supervisors will agree that salary increases Should be tied directly
to performance appraisal.

75
Sub-hypothesis 3, B:

Subordinates will feel that salary increases Should not be tied
directly to performance appraisal. -

These hypotheses were accepted and rejected, respectively.
Sub-hypothesis 3, A was supported by the responses to Question #12, sub-
hypothesis 3, B was not supported by the same responses.

Sub-hypothesis 3, C:

There will be a Significant difference in the rank order of the
purposes Of performance appraisal for supervisors and subordinates.

This hypothesis was not supported by responses to Question #13,
positive correlation point eight-eight-five-seven (.8857) (significance

level, point zero-one-zero (.010), and therefore was rejected.

CHAPTER V

SUMMARY, CONCLUSIONS , AND

IMPLICATIONS

Introduction

 

Performance appraisal, whether formalized or not, exists in all
organizations. The question is not whether it does, or should exist,
but, rather, what form it should take.

Its purposes are generally assumed to be (a) to improve employ-
ee performance through coaching and counseling, (b) to provide management
with information to support personnel decisions, and (c) to let the em-
ployee know where he stands relative to the organization's expectations
regarding job performance.

Chapter I

Although performance appraisal is widely used, considerable
support was found in the literature to indicate that there is disagree-
ment as to (a) its purpose, (b) its relative importance, and (c) the form
it should take.

Evidence indicated that there appeared to be disagreement on
these points, especially between superiors and subordinates, even though
they seemed to be in substantial agreement that performance appraisal
was necessary.

The purpose Of this study was to substantiate that, in fact,
differences did exist in the perceptions Of performance appraisal by

76

77

superior and subordinate in regard to purpose, need, and method.

The following hypotheses were developed to fulfill this pur-

pose:

Hypothesis 1:

 

The need for performance appraisal will be seen as more important
by supervisors than subordinates.

Hypothesis 2:

 

The preferred method for carrying out a perfOrmance appraisal will
be perceived differently by supervisors and subordinates.

A.

B.

C.

D.

Supervisors will prefer a method which maximizes their
input.

Subordinates will prefer a method which maximizes their
inPUt o

Supervisors will prefer a method which gives more weight
to subjective criteria.

Subordinates will prefer a method which gives more weight
to Objective criteria.

Hypothesis 3:

 

The pepposes Of performance appraisal will be perceived differently
by supervisors and subordinates.

A.

B.

C.

Chapter II

 

Supervisors will agree that salary increases Should be
tied directly to performance appraisal.

Subordinates will feel that salary increases should not
be tied directly to performance appraisal.

There will be a significant difference in the rank order
Of the purposes Of performance appraisal for supervisors
and subordinates.

In Chapter II, the literature on performance appraisal was re-

viewed; opinions on the need for it were examined, problems with it were

considered, and recommendations for dealing with it were presented.

Additionally, a representative variety of performance appraisal methods

78
were reviewed, along with a summary Of their relative strengths and weak-
nesses. Finally, factors which affect performance appraisal were report-
ed.

Although Opinion appeared to be divided regarding the merits of
performance appraisal, most authorities, management, and even subordin-
ates, seemed tO feel that performance appraisal was necessary, even
though they Often disagreed on its purpose, and in spite of its failures.

Considerable evidence was fOund to indicate that many of the
problems with performance appraisal result from the fact that such an
appraisal is emotionally-laden. Performance appraisal appears weakest
where it is most-needed, in dealing with the employee who is performing
below standard.

Other problems result from personal biases, inconsistencies,
and lack Of training on the part of the evaluator or from confusion on
the part Of management regarding the purpose of performance appraisal.

Despite these problems, performance appraisal continues to be
used by most organizations and much time and effort have been spent to
make it work more productively.

Some suggestions that have proven fruitful are to keep the
appraisal system simple, clearly defined, and participative in nature.

It is important that the superior and subordinate view performance
appraisal as an opportunity for growth and not an authoritarian critique.
Some additional approaches that Show promise are peer appraisal and

training for evaluators.

Chapter III

The research methodology and study design were presented in

Chapter III. A survey questionnaire, developed by the researcher in

79
consultation with experts in the field, was administered to the full-
time home Office employees (N = 565) of an internationally-known corpora-
tion headquartered in the Midwest. Three hundred and eighty-eight (388)
questionnaires or sixty-nine percent (69%) were returned by the cut-Off
date.

Chapter IV

 

The data from the questionnaires was analyzed in the following
manner:

1) Questions 9, 10, 11, 12 and 14 were analyzed by carrying out
the chi square test, comparing the responses of superiors and
subordinates.

2) Question 14 was analyzed by computing the Spearman correlation
coefficient for the responses Of superiors and subordinates.

3) Cross-tabulations were carried out on the responses of the
independent variable groups Of age, sex, race, and level Of
education to questions 9 through 14. The independent variable
groups were determined by the responses to questions 1 through
8.

Conclusions

 

On the basis of the application Of these statistical tests to
the data received, the following hypotheses were accepted:

Hypothesis 1:

 

The need for performance appraisal will be seen as more important
by supervisors than subordinates.

Hypothesis 2:

The preferred method for carrying out a perfOrmance appraisal will
be perceived differently by supervisors than subordinates.

A. Supervisors will prefer a method which maximizes their
input.

B. Subordinates will prefer a method which maximizes their

input.
Hypothesis 3:

 

A. Supervisors will agree that salary increases should be tied
directly to performance appraisal.

80
On the basis Of the application of the previously-discussed
statistical tests to the data received, the following hypotheses were
rejected:

Hypothesis 2:

 

B. Subordinates will feel that salary increases Should not be
tied directly to performance appraisal.

C. There will be a Significant difference in the rank order Of
the purposes of performance appraisal for supervisors and
subordinates.

Hypothesis 3:

 

The peppose Of performance appraisal will be perceived differently
by supervisors and subordinates.

B. Supervisors will prefer a method which gives more weight to
subjective criteria.

C. Subordinates will prefer a method which gives more weight
to Objective criteria.

The results Of this study substantiate the position taken by
authors such as Conant139, Meyerl40, and Meyer, Kay, and Frenchl4l, that
the need for performance appraisal will be seen as more important by the
superior than by the subordinate. It also provides support for studies
by Swerdlow142, Connollyl43, Blackburn and Clarkl44, and Williamsl45,

that the preferred method for performance appraisal is perceived

 

139Conant, op.cit., p. 77.

140Meyer, op.cit., p. 508.

141Meyer, Kay, and French, op.cit., p. 123.

142Swerdlow, op.cit.

143Connolly, op.cit.

144Blackburn and Clark, op.cit., pp. 242-256.

145Williams, op.cit., p. 89.

81
differently by superior than subordinate.

The outcomes of this study did not support the position taken
by authors such as Richards146 and Kellogg147, who believe that the pur-
poses Of perfOrmance appraisal are seen differently by superiors and
subordinates.

The results Of this study also emphasized the preference Of
most superiors and subordinates for an appraisal system which allows for
a self-evaluation by the ratee. This agrees with recommendations by
Patzl48, Grellerl49, and Frenchlso.

Objective appraisal criteria were not seen as important,
especially to subordinates, as had been expected. Most reSpondentS
seemed to favor a system which allowed for at least some subjective
criteria as well.

Respondents in this study seemed to feel that salary increases
should be closely tied to performance appraisals.

Educational level appeared tO have an effect upon an indivi-
dual's perception Of performance appraisal. It was found that the more-
educated persons favored more frequent performance appraisals, and that

they gave greater weight to Objective criteria in performance appraisal

than did those less-educated.

 

146Richards, op.cit., p. 560.

147Kellogg, op.cit., p. 90.

148Patz, op.cit., p. 80

149Greller, op.cit., pp. 544-9.

150French, et.a1., op.cit., pp. 3-20.

82
Sex also had an effect, with more women preferring a system
which allowed for ratee input than men. In addition, women gave greater
importance in performance appraisal purposes to improving communication
and determining salary than men, who rated professional growth higher
than either of these items.

Inferences And Recommendations

 

In applying the results of this study and literature review to
the corporation in which the study was done, the following recommendations
are made:

1. The company should consider development Of a performance review
system which allows for appraisee input (self-evaluation).

2. Consideration might be given to developing an appraisal system
for salaried-exempt employees which relies less heavily upon an
.MBO format.

3. The company should be aware of the preference of more-educated
personnel for an appraisal system weighted more heavily toward
Objective criteria, the reverse being true for less-educated
personnel. Mest employees, however, prefer a system utilizing
at least some subjective criteria.

4. Thought could be given to the advisability Of developing differ-
ent appraisal systems to meet different purposes and for differ-
ent departments.

5. Consideration should be given to the fact that women (as indicated
in the review of the literature and substantiated in this study)
perceive performance appraisal differently in a number of areas

than do men.

83

Implications For Future Research

 

One area in which this study was unable to Obtain any meaning-
ful data was in regard to the effect Of ethnic group or race upon per-
formance appraisal, largely because Of sample make-up. This area, it
seems, could be profitably investigated.

It should also be borne in mind that a large proportion of this
sample was highly educated and, as a result, the findings may not be
applicable to a less-educated population.

This study did not lend support to the related literature
which indicated that the purposes of performance appraisal were seen
differently by superior and subordinate. Two factors that may have
affected the findings in this area are (a) that the company in the study
had used performance appraisal for a considerable period of time, a con-
dition which may have had a "leveling" effect regarding perceptions of
performance appraisal by superiors and subordinates, and (b) the rela-
tively high level Of education Of the respondents may have had a similar
effect.

NO attempt was made in this study to evaluate a particular
appraisal system. This may prove to be a fruitful area Of enquiry, for
most Of the literature and research deals more with the effects upon
reliability and validity Of a given type Of system than the feelings of
the individuals who use it.

Neither did this study examine the effect Of occupational type
upon an individual's perception of performance appraisal. It is possible,
that this variable, too, could prove worthy of research.

Overall, this writer feels that considerably more research needs

to be done on the "human factors" in performance appraisal. Since it is

84

an emotionally-charged process, it would appear profitable for more
research to be directed toward investigating the perceptions of superior
and subordinate toward these human factors --- areas such as what type
Of system individuals prefer and why --- how the appraisal interview
can be "defused" --- why, in spite of its shortcomings, both superior
and subordinate feel that performance appraisal is necessary. Another
area Of investigation might lie in studying the effectiveness Of some
approaches advocated by various authors such as the "varimat" and bi-
monthly appraisals. Research needs to be directed, too, toward learn-
ing why behaviorally-anchored rating scales, which appear so promising,
have yet to live up to that promise.

Overall, one must conclude that there is still much to learn
about performance appraisal, in spite Of all the work that has been done
to date. For a tOOl that is so widely used and upon whose use SO many

decisions are made, we know far too little about it.

APPENDICES

APPENDIX A

COVER LETTER

APPENDIX A

COVER LETTER

November 14, 1977

Dear Respondent:

I would greatly appreciate it if you could take a little time to
fill out the attached questionnaire for me and mail it in the attached,
stamped, self-addressed envelope within the next few days. Preliminary
tests indicate that it will take no more than five to ten minutes to
complete. Your participation, Of course, is completely voluntary.

I am conducting research at Michigan State university and this
survey form is part Of my research. I have no connection with xxxxxx
and am not carrying out this survey at their request or upon their
behalf. xxxxxx management has, however, allowed me tO carry out my
survey here. I would want you to understand that the results Of this
survey have nothing to do with any xxxxxx performance review system
nor are they intended to bring about or influence any changes in the
present xxxxxx performance review system.

It would be helpful if you would fill out the questionnaire in
terms Of what you would like to see in performance review, rather than
in regard to your past experience with performance review.

Please be assured that your reSponse will be completely anonymous.
Please do not put your name, job, or department on the form.

Thank you very much for your help!

Very truly yours,

We?

HAROLD R.

85

APPENDIX B

QUESTIONNAIRE

APPENDIX B

QUESTIONNAIRE
l. I am a supervisor/manager yes no
2. I conduct performance reviews with subordinates yes no

 

3. I am salary exempt salary non-exempt

4. Years at present job level 0-5 "ll-15 over 20
6-10 16-20

5. Your age under 20 30-39 50-59

20-29 40-49 60 and over
6. Your education some high school Bachelor's Degree
high school grad Master's Degree
some college Doctorate
7. Are you male female
8. Your Ethnic group Caucasian Chicano/Latino
Black other

Performance appraisal may be defined as an appraisal or evaluation of
an individual's performance of his/her job functions.

9. Some type of performance appraisal Should be carried out for all

employees.
strongly agree disagree
agree strongly disagree

10. If you agree, how Often should this appraisal be carried out?
twice a year every two years

annually other (specify)

 

 

86

87
11. A performance appraisal should be done by: (check one)

the immediate supervisor only

the immediate supervisor and the person supervised (self-
evaluation)

the department head only
the department head and the person supervised (self-evaluation)
the department head and the immediate supervisor

the person being evaluated only (self-evaluation)

 

a person from outside the company (example, consultant)

 

the department head, immediate supervisor and person supervised

(self-evaluation)

other (specify)

 

 

 

12. An individual's salary increase should be tied directly to his/her
performance appraisal.

strongly agree disagree
agree strongly disagree
13. Listed below are 6 possible purposes fOr or uses Of performance

appraisal. Please rank them in order Of importance as you see them
with 1 being highest and 6 lowest.

provide feedback to employee on job performance

assist in employee's professional growth

determine salary merit increases

determine promotion within organization

provide an ongoing record Of employee's growth and development

improve communications between supervisor and employee

14. Performance appraisal can be carried out using Objective and/or

subjective criteria. Examples Of Objective criteria might be:
quantity of work done, Objectives met, etc. Examples Of subjective
criteria might be: employee attitudes, personality traits, quality

of work done, etc. In your Opinion, performance appraisal should be
based on: (check one)

 

 

____pbjective criteria only

____subjective criteria only

____primari1y Objective, but include some subjective criteria
_primarily subjective, but include some Obj ective criteria
____equally based on both subjective and Objective criteria

SELECTED BIBLIOGRAPHY

SELECTED BIBLIOGRAPHY

Baird, Lloyd S. "Self and Superior Ratings Of Performance: As Related
to Self-Esteem and Satisfaction with Supervision," Academy of
Management JOurnal, Vol. 20, NO. 2, June 1977, pp. 291-300.

 

 

Bass, Alan R. and Turner, John H. "Ethnic Group Differences in Rela-
tionships Among Criteria of Job Performance," Journal Of
Applied Psyehology, VOl. 57, April 1973, pp. 101-9.

 

 

Bassett, Glenn A. The New Face of Communication. Second Edition, New
York: American Management Assoc., 1968.

 

Bayroff, A. G.; Haggerty, Helen R.; and Rundquist, E. A. "Validity Of
Ratings as Related to Rating Techniques and Conditions,"
appearing in Whisler, Thomas L. and Harper, Shirley F.
Performance Appraisal, New York: Holt Rinehart & Winston,
1962, pp. 112-125.

 

Beer, Michael and Ruh, Robert A. "Employee Growth Through Performance
Management," Harvard Business Review, Vol. 54, July 1976,
pp. 59-66.

 

Bernardin, H. John; LaShells, Mary B.; Smith, Patricia C.; and Alvares,
Kenneth M. "Behavioral Expectation Scales: Effects of
Developmental Procedures and Formats," JOurnal Of Applied
Psychology, VOl. 61, February 1976, pp. 75-9.

 

Bigoness, William J. "Effect Of Applicant's Sex, Race and Performance
on Employer's Ratings: Some Additional Findings," Journal Of
Applied Psychology, Vol. 61, February 1976, pp. 80-4.

 

Blackburn, Robert T. and Clark, Mary JO. "An Assessment Of Faculty Per-
formance: Some Correlates Between Administrator, Colleague,
Student, and Self-Ratings," Sociology Of Education, Vol. 48,
Spring 1975, pp. 242-56.

 

Blumenfeld, Warren S. and Janus, Sidney Q. "Interrater Reliability Of a
Performance Criterion for a very Homogeneous Group of Managers,"
Psychological Reports, VOl. 35, December 1974, pp. 70-76.

Borman, Walter C. "Behavior-Based vs. Trait-Oriented Performance Ratings:

An Empirical Study," Journal of Applied Psychology, Vol. 60,
October 1975, pp. 561-5.

88

89

Borman, Walter C. "Effects of Instructions to Avoid Halo Error on
Reliability and Validity Of Performance Evaluation Ratings,"
Journal Of Applied Psychology, Vol. 60, October 1975,
pp. 556-60.

 

Broom, Leonard and Selznick, Philip. Sociology. Fourth Edition, New
York: Harper & Row, 1968.

Conant, James C. "The Performance Appraisal," Business Horizons, Vol.
XVI, NO. 3, June 1973, pp. 73-8.

 

Connolly, Eugene F. The Evaluation of Massachusetts Community College
English Teachers: Comparisons of Perceptions Of Chairmen and
Faculty. Doctoral Dissertation, Boston College, 1975.

 

Dow Corning, Guide in Reviewing Professional Employees, November 1976,
Dow Corning Corp.

Duffy, Kirt E. and Webber, Robert E. "On 'Relative' Rating Systems,"
Personnel Psychology, Vol. 27, Summer 1974, pp. 307-11.

 

Feild, Hubert S. and Holley, William H. "Subordinates' Characteristics,
Supervisors' Ratings, and Decisions TO Discuss Appraisal
Results," Academy of Management Journal, Vol. 20, NO. 2, June
1977, pp. 315-21.

Finley, David Merle. The Effects Of Scale Continuity And Behavioral
Anchor Specificity Upon The Psyehometric Properties Of Perfor-
mance Rating Scales. Doctoral Dissertation, university of
Houston, 1975.

Ford, Robert C. and Jennings, Kenneth M. "HOw TO Make Performance
Appraisals More Effective," Personnel, Vol. 54, March-April
1977, pp. 51-6.

French, John R. P.; Kay, Emmanuel; and Meyer, Herbert H. "Participation
and the Appraisal System," Human Relations, Vol. 19, February
1966, pp. 3-20.

 

Goodale, James G. and Burke, Ronald J. "Behaviorally-Based Rating Scales
Need Not Be Job Specific," Journal of Applied Psychology,
Vol. 60, June 1975, pp. 389-91.

Greller, Martin M. "Subordinate Participation and Reactions tO the

Appraisal Interview," JOurnal Of Applied Psychology, Vol. 60,
October 1975, pp. 544-9.

Grey, Ronald J. and Kipnis, David. "Untangling The Performance Appraisal
Dilemma: The Influence Of Perceived Organizational Context On
The Evaluative Process," Journal of Applied Psychology, Vol.
61, June 1976, pp. 329-35.

9O

Hillery, Joseph ML and Wexley, Kenneth N. "Participation Effects In
Appraisal Interviews Conducted In a Training Situation,"
Journal Of Applied Psychology, Vol. 59, April 1974, pp. 168-71.

 

Huber, Norman Andrew. Superior-Subordinate Similarity, Performance
Evaluation And Job Satisfaction. Doctoral Dissertation, Wayne
State university, 1970.

 

Jordan, Jackie B. The Development And Evaluation Of Behaviorally
Anchored Rating Scales For The Assessment Of Department Chair-
man Performance. Doctoral Dissertation, University of
Kentucky, 1976.

 

 

Keeley, Michael, "Subjective Performance Evaluation And Person-Role
Conflict under Conditions Of Uncertainty," Academy of Manage-
ment JOurnal, Vol. 20, NO. 2, June 1977, pp. 301-14.

 

 

Kellogg, Marion S. What TO DO About Performance Appraisal. New York:
AMACOM, 1975.

Kellogg, Marion 5. "New Angles in Appraisal," appearing in Whisler,
Thomas L. and Harper, Shirley F. Performance Appraisal. New
YOrk: Holt Rinehart & Winston, 1962, pp. 88-95.

 

Kluckhohn, Clyde. Culture And Behavior. New York: The Free Press of
Glencoe, 1962.

Krant, Allen 1. "Prediction Of Managerial Success By Peer And Training
Staff Ratings," Journal of Applied Psychology, Vol. 60,
February 1975, pp. 14-19.

LaBean, James H. Employee Perceived Values Of Job Performance Reviews.
Masters Thesis, Central Michigan University, 1975.

Latham, Gary P.; wexley, Kenneth N.; Pursell, Elliott D. "Training
Managers TO Minimize Rating Errors In The Observation Of
Behavior," Journal Of Applied Psychology, Vol. 60, October
1975, pp. 550-5.

Lawshe, C. H.; Kephart, N. C.; and McCormick, E. J. "The Paired
Comparison Technique For Rating Performance Of Industrial
Employees," appearing in Whisler, Thomas L. and Harper,
Shirley F. Performance Appraisal. New York: Holt Rinehart
& Winston, 1962, pp. 209-15.

Lee, M. Blaine and Zwerman, William L. "Designing a Motivating and
Team Building Employee Appraisal System," Personnel Journal,
Vol. 55, July 1976, pp. 354-7.

Levinson, Harry. "Appraisal Of What Performance," Harvard Business
Review, Vol. 54, July 1976, pp. 30-2.

 

91

Lewin, Arie Y. and Zwany, Abram. "Peer Nominations: A Model, Literature
Critique And A Paradigm For Research," Personnel Psychology,
Vol. 29, Autumn 1976, pp. 423-47.

 

Locher, Alan H. and Teel, Kenneth 5. "Performance Appraisal - A Survey
Of Current Practices," Personnel Journal, Vol. 56, May 1977,
pp. 245-7.

 

McAfee, Bruce and Green, Blake. "Selecting A Performance Appraisal
Method," Personnel Administrator, Vol. 22, June 1977, pp. 53-8.

 

McGregor, Douglas. Leadership And Motivation. Massachusetts: M.I.T.
Press, 1966.

 

Meyer, Herbert H. "The Annual Performance Review Discussion -- Making
It Constructive," Personnel Journal, Vol. 56, October 1977,
pp. 508-11.

 

Meyer, Herbert H.; Kay, Emmanuel; and French, JOhn R. P. "Split Roles
In Performance Appraisal," Harvard Business Review, January-
February 1965, p. 123.

 

Millard, Cheedle William. The Development And Evaluation Of Behavioral
Criteria For MeasuripgiThe Performance Of Non-Operational

Employees. Doctoral Dissertation, University of Nebraska,
1974.

Miner, John B. Personnel and Industrial Relations. New York: The
Macmillan Co., 1969.

Morrisey, George L. Appraisal and Development Through Objectives and
Results. Reading, Mass.: Addison-Wesley Publishing Co.,
1972.

Myers, M. Scott. Every Employer A Manager. New York: McGraw-Hill Book
Co., 1970.

Oberg, Winston. "Make Performance Appraisal Relevant," Harvard Business
‘ Review, Vol. 50, January 1972, pp. 61-7.

 

Odiorne, George S. Personnel Policy: Issues and Practices. Columbus,
0.: Chas. E. Merrill Books, 1963.

Patz, Alan. "Performance Appraisal, Useful But Still Resisted," Harvard
Business Review, Vol. 53, May 1975, pp. 74-80.

Quinn, James Lester. Performance Appraisals: The Relation Between
Rating§_And Selected Characteristics Of The Rating_Dyads.
Doctoral Dissertation, Indiana university, 1967.

92

Richards, Kenneth. "A New Concept In Performance Appraisal," appearing
in Whisler, Thomas L. and Harper, Shirley F. Performance
Appraisal. New York: Holt Rinehart & Winston, 1962,
pp. 549-65.

 

Rieder, George A. "Performance Review - A Mixed Bag," Harvard Business
Review, Vol. 51, July 1973, pp. 61-7.

 

Ruch,_Floyd L. Psychology And Life. New York: Scott, Foresman & Co.,
1953. ’

 

Ryan, Thomas A. "Merit Rating Criticized," appearing in Whisler,
Thomas L. and Harper, Shirley F. Performance Appraisal. New
York: Holt Rinehart & Winston, 1962, pp. 161-9.

 

Salton, Gary J. "VARIMAT: Variable Formal Performance Appraisal,"
Personnel Administrator, Vol. 22, June 1977, pp. 53-8.

 

Schwab, Donald P.; Heneman, Herbert G. III; and DeCotiis, Thomas A.
"Behaviorally Anchored Rating Scales: A Review Of The Litera-
ture," Personnel Psychology, Vol. 28, Winter 1975, pp. 549-62.

 

Sloan, Stanley and Johnson, Alton C. "New Context Of Personnel
Appraisal," Harvard Business Review, Vol. 46, November 1968,
pp. 14-20.

 

Sofer, C. and Tuchman, M. "Appraisal Interviews and the Structure of
Colleague Relations," Sociology Review, Vol. 18, November
1970, pp. 365-91.

 

Summers, Basil Irvin. Exploratory Study: Comparison Of 'Cognitive
Styles' TO Employee Evaluation And Employee Satisfaction.
Doctoral Dissertation, Wayne State university, 1971.

Swerdlow, Robert A. An Investigation And Evaluation Of Measures Of
Performance Of Executives Relating TO Four Types Of Retail
Operations. Doctoral Dissertation, university of Arkansas,
1976.

Thompson, Paul H. and Dalton, Gene W. "Performance Appraisal: Managers
Beware," Harvard Business Review, Vol. 48, January 1970,
pp. 149-57.

Van Duren, Allan. The Value Of Performance Appraisal In The Motivation
Of Engineering Personnel. Doctoral Dissertation, United States
International University, 1976.

Whisler, Thomas L. and Harper, Shirley F. Performance Appraisal. New
York: Holt Rinehart & Winston, 1962.

White, B. Frank and Barnes, Louis B. "Power Networks in the Appraisal
Process," Harvard Business Review, Vol. 49, May 1971, pp. 101-9.

93

Whitla, Dean K. and Tirrell, JOhn E. "The Validity Of Ratings of
Several Levels Of Supervisors," appearing in Whisler, Thomas
L. and Harper, Shirley F. Performance Appraisal. New York:
Holt Rinehart & Winston, 1962, pp. 132-5.

Williams, M. R. Performance Appraisal in Management. London: William
Heineman LTD., 1972.

Zedeck, Sheldon; Imparato, Nicholas; Krausz, Meshe; Oleno, Timothy.
"Development of Behaviorally-Anchored Rating Scales as a
Function Of Organizational Level," JOurnal Of Applied
Psychology, Vol. 59, April 1974, pp. 249-52.