THE PERSONNEL ASSESSMENT CENTER: A STUDY
OF EFFECTS UPON ASSESSEES

Thesis for the, Degree of Ph. D. '
MICHTGAN STATE UNIVERSITY
DAT/TD SELLERS VCGELS, Jr.

1973

 

 

 

This is to certify that the

thesis entitled

THE PERSONNEL ASSESSMENT CENTER:
A STUDY OF EFFECTS UPON ASSESSEES

presented by

David Sellers Vogels, Jr.

has been accepted towards fulfillment *
of the requirements for ‘

Ph . D degree in Business Adminis trat ion

 

/ £6 ‘Lq/ ./o‘5(

r flprofessor

Date May 18, 1973 l

 

0-7 639

 

ABSTRACT

THE PERSONNEL ASSESSMENT CENTER:
A STUDY OF EFFECTS UPON ASSESSEES

BY
David Sellers Vogels, Jr.

Within the past few years, many organizations have
turned to the personnel assessment center as a method for
identifying management potential. Much research has been
devoted to ascertaining the predictive validity of this
method. Most of the research evidence indicates that,
with varying degrees of validity, prediction of useful
criteria can be made from the multiple evaluation infor-
mation obtained at a personnel assessment center.

However, desPite the ever-increasing use of personnel
assessment techniques and methods, virtually no research
has been devoted to determining the existence or extent of
any impact upon the assessee arising from his personnel
assessment center experience. The present research has
been concerned primarily with the effects of a personnel
assessment exPerience upon the assessee's job performance
and job satisfaction after assessment.

It is logical to assume that assessee reaction to

David Sellers Vogels, Jr.
assessment will differ and that these differences might be
accounted for at least in part by personality character-
istics, particularly self-esteem. This logical assumption
led to the develoPment of the hypotheses. Basically these
state that after an assessment center experience, job per-
formance and job satisfaction would:

1. Increase for those assessees with high self-
esteem who attained above median assessment
ratings.

2. Not change for those assessees with high self-
esteem who attained below median assessment
ratings.

3. Not change for those assessees with low self-
esteem who attained above median assessment
ratings.

4. Decrease for those assessees with low self-
esteem who attained below median assessment
ratings.

The subjects were 60 managers who participated in a
personnel assessment center, conducted by the parts divi-
sion of a manufacturing firm in the automotive industry.
As part of the assessment process, assessment ratings were
obtained on ten variables for each subject. These were
combined for use as a dependent variable here. Other data
were collected as part of the research project. These
were measures of:

1. Personality characteristics, where the measuring

instrument used was the Ghiselli Self-Descrip-
tion Inventory (SDI). The SDI was administered

as part of the assessment center battery.

2. Job performance, where a supervisor's rating
was used. A rating was obtained on the assessee

David Sellers Vogels, Jr.

before assessment and again about six months
after assessment.

3. Job satisfaction, where the measuring instru-
ment used was the Job Descriptive Index (JDI).
The JDI was administered as part of the assess-
ment center battery and again about six months
after assessment.

To test the hypotheses, measures of job performance
and job satisfaction measures were examined before assess-
ment and six months after assessment. A three-way analysis
of variance was used. That is, a 2 x 2 x 2 factorial de-

sign was employed. In the design, the main effects were:

(1) assessment rating; (2) level of self-esteem; and

 

(3) time. The dependent variables were job performance

 

and job satisfaction. Where the analysis indicated a sig-
nificant F ratio, the Scheffe method of means comparison
was used to test the significance between all possible com-
binations of pairs of means.

As a result of the hypothesis testing, Hypotheses 1
and 4 were rejected and Hypothesis 3 was accepted. ’
Hypothesis 2 was accepted in part since job performance
did not change; and it was rejected in part since job
satisfaction with promotions decreased.

As just indicated only one significant change, which
was contrary to the hypothesized direction, was found:
satisfaction with promotions declined for those assessees
with high-self esteem who attained belOw median assessment

ratings. This finding suggests that those who think highly

David Sellers Vogels, Jr.
of themselves may be disturbed by their low assessment
rating or "failure". They then apparently externalize
this failure to the promotion system, a system which may
be associated in their minds with the assessment program.

If the research findings are replicated in other
organizations, supervisors who have pondered the impact
of assessment ratings on job performance and job satis-
faction may be reassured that the only area of concern
seems to be promotion satisfaction. Other than that,

there is no apparent impact six months after assessment.

THE PERSONNEL ASSESSMENT CENTER:

A STUDY OF EFFECTS UPON ASSESSEES

BY

David Sellers Vogels, Jr.

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Management

1973

ACKNOWLEDGMENTS

It is with a deep sense of gratitude that I acknowl—
edge the assistance received during the course of my
doctoral studies. While many peOple certainly warrant
a "thank you", I believe that I must particularly single
out several groups of peOple for Special thanks because
of the support and encouragement they gave me.

I thank the personnel at the Air Force Institute of
Technology for giving me the Opportunity to undertake a
doctoral program in business administration. Colonel
Robert H. McIntire, Director of Civilian Institutions,
always stood ready to assist me accomplish the various
duties I had by virtue of being the senior Air Force
student on campus. Major James S. Austin, Jr., Program
Manager, rendered much invaluable support and advice as
he guided me through my three-year academic tour.

I thank the personnel of Detachment 380, Air Force
ROTC for their fine administrative support during my
three years at Michigan State University. Colonel LeRoy

M. Wenstrom, Professor of Aerospace Studies, rendered

ii

vast personal support and relieved me of many ”paper-
work" burdens; thereby, enabling me to devote more atten-
tion to my student tasks.

I thank Mrs. Josephine McKenzie for her excellent
service in typing this thesis under the severe pressure
of time. Her skill made the final "wrap-up” of the
thesis an accomplished task.

I thank my thesis committee for their ever-ready
assistance and guidance. Professor Thomas H. Patten,
Jr., committee member, with his warm personal friendship
sustained me through some of the troubled times I experi—
enced during my doctoral program. Professor Frederic R.
Wickert, committee member, with his many constructive
comments saved me from several potential ”traps” I had
set for myself during the writing of the thesis. Pro-
fessor Henry L. Tosi, committee chairman, with his un-
stinting efforts on my behalf, literally carried me
through my entire doctoral program. To Dr. Tosi I must
give a special "thank you", because his energy, zeal,
ready availability, and constant attention to my thesis
really made it possible for me to ultimately attain the
PhD degree.

I thank my family for their many sacrifices in my
behalf. My sons David, Bob, Tad, and Jon have never

iii

enjoyed having a father who was not a student in some pro-
gram of academic endeavor. But despite the many drawbacks
being a student placces on one's family life, they have
never complained. My wife Mimi through over twenty years
of our married life has borne a heavy burden as I have
pursued my personal objectives for advanced education.
While at times she may have wondered about the validity
of my goals, she never hesitated in backing my desire to
achieve a PhD degree. To her I owe a very special "thank
you". It is to her that I dedicate this thesis for her
patience, understanding, encouragement, faith, and love

which ultimately made it all come true.

iv

TABLE OF CONTENTS

LIST OF TABLES. . . . . . . . . . . . . . . . . . .

LIST OF FIGURES . . . . . . . . . . . . . . . . . .

TABLE OF APPENDICES . . . . . . . . . . . . . . . .

Chapter
1

THE PERSONNEL ASSESSMENT CENTER. . . . . .

Introduction . . . . . . . . . . . . . .
DeveIOpment of Assessment Centers. . . .
Typical Assessment Center. . . . . . . .
Research on Assessment Centers . . . . .
Research Objective . . . . . . . . . . .

RESEARCH DESIGN AND METHODOLOGY. . . . . .

Theoretical Concepts . . . . . . . . . .
Model and Hypotheses . . . . . . . . . .
Research Design. . . . . . . . . . . .

Research Sample. . . . . . . . . . . . .
Measuring Instruments. . . . . . . . . .
Statistical Methods. . . . . . . . . . .
Summary. . . . . . . . . . . . . . . . .

RESEARCH RESULTS . . . . . . . . . . . . .

The Impact of Assessment on Performance
and Satisfaction. . . . . . . . . . . .
Relationships of Ghiselli SDI Traits . .
Summary. . . . . . . . . . . . . . . . .

SUMMARY, DISCUSSION, AND CONCLUSIONS . . .

The Impact of Assessment on Performance
and Satisfaction. . . . . . . . . . . .

Relationships of Ghiselli SDI Traits .
Implications for Management. . . . . . .

V

Page
vii

ix

10
16
25

31

31
39
42
48
56
7O
72

74
75
85
91
92
95

106
110

Chapter Page

Suggestions for Future Research. . . . . . 112
Conclusion . . . . . . . . . . . . . . .‘. 113
LIST OF REFERENCES. . . . . . . . . . . . . . . . . . 115

APPENDICES. . . . . . . . . . . . . . . . . . . . . . 121

vi

Table

2-4

2-8

LIST OF TABLES

Subjects Involved in Analysis of Variance

of Various Effects of Assessment . . .

Subjects Involved in Analyses of Ghiselli

Characteristics and Various Criterion.

Coefficients of Correlation Between the

Scores of Managers, Supervisors, and
Workers on the Various SDI Scales and
Their JOb Success. . . . . . . . . . .

Assessment Rating Scale Inter—Item
Correlations . . . . . . . . . . . . .

Reliabilities and Validities of JDI
Scales . . . . . . . . . . . . . . . .

Self-Rating Performance Rating Scale
Inter—Item Correlations. . . . . . . .

Supervisor Performance Rating Scale
Inter-Item Correlations. . . . . . . .

Peer-Average Performance Rating Scale
Inter-Item Correlations. . . . . . . .

Intercorrelations Among Assessment
Rating and Various Job Performance
Ratings I O O O O O O O O O O O O O O 0

Relationship Between: (A) Before and
After Assessment, (B) Supervisor Job
Performance Rating, and (C) Assessment
Rating for Assessees at Different
Levels of Self—Esteem. . . . . . . . .

vii

Page

49

49

59

62

64

66

67

68

69

76

Table Page

3-2 Relationship Between: (A) Before and
After Assessment, (B) Job Satisfaction
(Work), and (C) Assessment Rating for
Assessees at Different Levels of
Self-Esteem. . . . . . . . . . . . . . . . . 78

3-3 Relationship Between: (A) Before and
After Assessment, (B) Job Satisfaction
(Supervision), and (C) Assessment Rating
for Assessees at Different Levels of
Self-Esteem. . . . . . . . . . . . . . . . . 80

3-4 Relationship Between: (A) Before and
After Assessment, (B) Job Satisfaction
(Promotions), and (C) Assessment Rating
for Assessees at Diferent Levels of
Self-Esteem. . . . . . . . . . . . . . . . . 82

3-5 Relationship Between: (A) Before and
After Assessment, (B) Job Satisfaction
(Co-Workers), and (C) Assessment Rating
for Assessees at Different Levels of
Self-Esteem. . . . . . . . . . . . . . . . . 84

3—6 Correlations of Ghiselli Traits with

Overall Job Performance Ratings. . . . . . . 86
3-7 Correlations of Ghiselli Traits with

Job Satisfaction Scores. . . . . . . . . . . 88
3-8 Correlations of Ghiselli Traits with

Assessment Ratings . . . . . . . . . . . . . 9O

viii

LIST OF FIGURES

OSS Assessment Staff's Principles

of Assessment. . .

Model of Assessment Effects Upon

an Assessee. . . .

Identification and Designation of

Assessee Groups. .

Schematic of Analysis of Variance.

Variables and Techniques of the

Assessment Center.

ix

page

40

44

46

52

TABLE OF APPENDICES

Appendix Page
A Ghiselli Self-Description Inventory (SDI). . 121
B Job Descriptive Index (JDI). . . . . . . . . 123
C Performance Self—Rating Form . . . . . . . . 127
D Supervisor/Co-worker Performance Rating
Form (Before Assessment) . . . . . . . . . . 133
E Supervisor Performance Rating Form
(After Assessment) . . . . . . . . . . . . . 139

CHAPTER 1
THE PERSONNEL ASSESSMENT CENTER

Introduction

In the United States today, most peOple are concerned
about the utilization of resources. This concern is often
focused on the environmental impact of resource manage-
ment, but other facets of resource management are not
entirely ignored. One such facet is the management of
human resources. For an indication of this particular
concern one need only consider the increase over the past
decade of federal-state legislation regulating employment
practices (e.g., in the area of civil rights). After such
consideration, it is apparent that American society is
indeed concerned with the management of human resources.

Accepting the presence of this societal concern, one

can understand why,

The key occupational group in an industrial
society is management. Effective direc-
tion of human efforts -- whether in the
public or private sectors of the economy ~-
is central to the wise and efficient
utilization of human and material resources
(Campbell, et a1, 1970, p. l).

1

2

Accordingly, as Dunnette (1971) asserts, there is a
chronic need for more and better managers. In an effort
to overcome this short supply, within the past few years
some industrial firms and government agencies have turned
to the personnel assessment center as a method of identi-
fying managerial talent. An assessment center is a place
where judgments are made about the managerial potential
or develOpmental needs of personnel in the organization.

'Assessments', in this context are the
pooled judgments of several specially
trained managers who use a variety of
criteria to evaluate a man's performance
as he goes through several different
'test' situations. Usually some paper—
and—pencil tests are also used, and an
intensive interview is a normal part of
the assessment procedure. It is this
matter of multiple judgments based upon
observations of performance in several
situations that is the crux of the assess-
ment center method (Wikstrom, 1967, p. 39).

It is estimated that more than one hundred organiza—
tions are using the assessment method for evaluating em—
ployee potential (Byham, 1971). Many thousands of indi—
viduals have been participants at an assessment center.
One estimate is that some 100,000 have been assessed
(Jaffee, et a1, 1970). The wide-Spread use of this
approach represents the latest stage in the develOpment

of the assessment method -- a method which first evolved

in the mid-1930's.

Development of Assessment Centers
The first use of the assessment method for evaluating
individuals is generally credited to Murray (Taft, 1959). He
used a series of interviews, tests, and experimental pro-
cedures which were devised for research in the area of the
psychology of personality. The research was conducted
during the period 1934-37 and involved mainly Harvard
male undergraduates who were paid to participate as sub-
jects. Assessors were members of the Harvard Psycho-
10gical Clinic, who rated assessees as a committee.
Murray credits these assessment procedures with
contributing to the study of personality in three ways:
1. A great deal of information is assembled
which can be used to interpret the re-
actions of each subject in each experiment.
In this manner, the experimentor is able
to discover many of the operating variables,
rather than having to content himself merely
with crude statistical results, such as are
obtained in most experiments.
2. Different aspects of personality are brought
to light by the different situations that are
presented.
3. Errors which arise from the experimentor's
personal vieWpoint are minimized (Murray,
1938, pp. 705—706).
World War II marked the beginning of the use of this

approach for selection purposes. Initially, German mili-

tary psychologists used the assessment method for

4
non-commissioned officer selection (Eysenck, 1953). In
1942, the British established War Office Selection Boards
(WOSBs) to select candidates for training leading to an
officer's commission in the British Army. The WOSB
assessment program involved a variety of standardized,
simulated real—life situations, interviews, and standard-
ized pencil-and-paper tests. The assessees lived at the
assessment centers during the program, which usually
lasted three days. Assessors were British Army officers,
psychologists, and psychiatrists. Judgments of the
assessors were pooled, with the president of the WOSB
rendering the final decision on selection (Morris, 1949).
A similar program was conducted by the Office of
Strategic Services (058) during 1944-45 to select candi—
dates for training as intelligence and espionage agents.
As with the WOSBs assessees, the candidates lived at the
assessment center for three days. During this period,
each assessee completed a number of paper-and-pencil
intelligence and personality tests and a detailed per-
sonal history questionnaire. Also assessees were given
two outdoor situational tests; an extensive personal
interview; and tests of prOpaganda skills, observation
and memory, and mechanical comprehension. In addition,

each assessee underwent a stress interview -- a procedure

5
designed to test the candidate's capacity to tolerate
severe emotional and intellectual strain.

Assessors were psychologists, psychiatrists, and
other social scientists. During the assessment program,
assessors independently rated the assessees. At the
completion of the prOgram their judgments were pooled,
for a staff decision on an assessee's final rating.

The OSS method of assessment was perhaps the first
to reflect an emerging emphasis on group and situational
exercises to assess the individual characteristics. This
emphasis on a man-group focus represented a shift from
earlier assessment methods which emphasized an individual-
personality focus.

Based upon their experience with assessment, the OSS
staff formulated a set of principles, which have provided
useful guidelines in establishing assessment centers

(Figure l—l).

FIGURE 1-1

088 ASSESSMENT STAFF'S PRINCIPLES
OF ASSESSMENT

 

 

1. - Make a preparatory analysis of all the jobs for which candidates
are to be assessed.

2. - On the basis of the preparatory analysis of jobs, list all the
personality determinants of success or failure in the perform-
ance of each job; and from this list select the variables to
be measured by the assessment process.

3. - Define (in words that are intelligible to the personnel officers
and administrators of the organization) a rating scale for each
personality variable on the selected list as well as for the one
over—all variable.

4. - Design a program of assessment procedures which will reveal the
strength of the selected variables.

4.1. — Plant the assessment procedures within a social matrix
composed of staff and candidates.

4.2. - Select several different types of procedures and several
procedures of the same type for estimating the strength
of each variable.

4.3. — Include in the program a number of situational tests in
which the candidate is required to function at the same
level of integration and under somewhat similar condi-
tions as he will be expected to function under in the
field.

5. - Construct a sufficient formulation of the personality of each
assessee before making Specific ratings, predictions, and
recommendations.

6. - Write, in nontechnical language, a personality sketch of each
assessee, which predictively describes him as a functioning
member of the organization.

7. - At the end of the assessment period hold a staff conference for
the purpose of reviewing and correcting the personality sketch
and of deciding on the ratings and recommendations of each
assessee.

8. - Construct eXperimental designs as frames for assessment proce-
dures so that all the data necessary for the solution of strate-
gic problems will be systematically obtained and recorded.

 

Source: OSS Assessment Staff, Assessment of Men (New York: Rinehart,
1948), pp. 28-56.

7

In the 1940's (1946-49), the Veterans Administration
conducted a clinical assessment program for the primary
purpose of validating certain assessment techniques.
Assessees were 471 clinical psychology graduate students,
who were rated by clinical and non-clinical psychologists
(Taft, 1959). While this assessment prOgram was not de-
signed to identify managerial potential, it did have an
impact on the kinds of assessment methods which are used
for this purpose today. For example, the evaluation of
the program led to the conclusion that both psychologists
and non-psychologists can competently act as assessors.
This finding is the basis for the wide-spread use of non-
psychologists as assessors in industrial assessment
centers. In general, the findings of this particular
research pregram with respect to validity of assessment
did much to foster confidence is this method of evaluating
the future potential of individuals in varying positions
(Kelly, 1951).

Several additional clinical assessment programs were
conducted in the early 1950's, again for the purpose of
validating various assessment techniques. Some of the
more important of these were: (1) the California Insti-
tute of Personality Assessment and Research (IPAR) program

assessing advanced graduate students: (2) the University

8
of Chicago program assessing students in theology, educa-
tion, and arts; (3) the Menninger School of Psychiatry
prOgram assessing psychiatric candidates; (4) the British
Civil Service Selection Boards (CISSB) program assessing
civil service candidates; and (5) the United States Air
Force program assessing Officer Candidate School (OCS)
applicants (Bray & Grant, 1966; Taft, 1959). These pro-
grams served to point out that the kind of assessment
center is quite different for different types of organ-
izations and organizational situations. For example, the
”buddy" rating technique used to assess officer candidates,
where individuals are in close and continuous contact
with one another for some 90 days, is not a technique
which could be readily used to assess individuals in an
industrial assessment program.

The American Telephone and Telegraph Company (AT&T)
in 1956 made the first industrial application of a per-
sonnel assessment center. It was used as a major research
methodology for its long-term Management Progress Study.
This study was intended to uncover information about the
personal deve10pment of men as they worked as managers
within the Bell System. As a part of the study, AT&T
operated assessment centers in the summers of 1956 through

1960 to Obtain information concerning the backgrounds and

9
abilities of individuals starting management careers in
the Bell System. The first assessees were young men,
recently recruited into the Michigan Bell Telephone Com-
pany (Wikstrom, 1967).

Managers in the firm were not permitted to use the
assessment reports in making promotion decisions. But
they did learn enough of the assessment procedures in-
volved to believe that these procedures would be useful
in selecting foremen from the skilled craftsmen working
for the company. So in 1958, acting upon this belief,
Michigan Bell opened an assessment center for the
appraisal of candidates for promotion to management from
vocational occupations. This was the first operational
assessment center, organized by an industrial firm, to
provide information to line managers (Wikstrom, 1967).
This assessment program -- the Personnel Assessment
Program -- is still in existence, assessing approximately
500 participants a year.

Currently, nineteen Bell System companies Operate 70
assessment centers and evaluate thousands of management
candidates each year. In addition, many organizations
both large and small have followed the lead of Michigan
Bell in establishing assessment centers. Some of these

organizations are Standard Oil (Ohio), J. C. Penney,

10
International Business Machines, General Electric, Sears,
Caterpillar Tractor, Olin-Mathieson, Ford Motor, Wolverine
Tube, Peace Corps, Internal Revenue Service, and Union
Carbide (Jaffee, et a1, 1970). Basically, these assess-
ment centers all follow a similar pattern. Hence, a
description of the "typical" assessment center will serve

as an adequate description of most assessment centers.

Typical Assessment Center

The basic purpose Of an organizational assessment
center is the identification of individuals with potential
for first-level supervisory positions in the organization.
A secondary purpose of assessment is the development of
individuals in the organization, enhancing their management
potential by providing important feedback so that they can
improve their effectiveness. That is, assessment results
are used to plan a program Of development for the assessee
to overcome weaknesses detected at the assessment center.
For example, the assessee may demonstrate difficulty in
expressing his ideas to others. To overcome this defi-
ciency,acourse in public Speaking or more participation in
group meetings could become part of the development plan

for this individual.

11

Assessees and Measures

Usually a small number of assessees, probably not
more than twelve, nominated by their supervisors as having
shown management potential through their job performance,
participate in an assessment session. A session lasts two
days, and during this time the assessee is subjected to a
wide variety of individual and group measurement tech-
niques. Individual measurement techniques evaluate the
individual without reference to his interaction with
others. Group measurements, on the other hand, evaluate
the individual as he interacts with others in a group
setting.

Individual measures typically include:

1. Interview. The most common interview situation
employed in assessment centers involves one
interviewer (an assessor) and one interviewee
(an assessee). The interview usually follows
a structured outline guide consisting Of a
number of carefully designed, Open-ended ques-
tions. These questions assure coverage of
desired subject areas, while providing the
interviewee an opportunity to respond freely
without structuring imposed by the question

itself.

12

2. In-Basket Exercise. The in-basket used as an
assessment technique usually contains such items
as reports, memoranda, letters, and other
materials requiring simulated action on the
part of the assessee. These materials are eSpe-
cially prepared to reflect a realistic Operating
situation to the individual being evaluated.

The task involved is to take the required actions
to deal with the problems presented by the mate-
rials, which are designed to present a wide
variety of problems with differing degrees of
complexity.

3. Psychological Tests. Various paper-and-pencil
tests are used in assessment, such as: School
and College Aptitude Test (SCAT), Contemporary
Affairs Test, and Strong Vocational Interest
Blank. Some projective tests such as the
Rorschach and the Thematic Apperception Test
(TAT) are used, but rather infrequently, in
assessment centers.

Group measures typically include:

1. Leaderless Group Discussion (LGD) Exercises. In

the LGD a half-dozen or more assessees discuss

some problem on which they all have approximately

Assessors

13
equal information. The problem is a controver—
sial one, so it lends itself to discussion and
interpersonal actions. Assessors rate the be-
havior of the assessees on a number of specific
characteristics which have been previously chosen
as important elements of the job for which the
candidates are being considered.
Management Games. Small group games provide a
live demonstration of factors such as the
assessee's ability to plan, communicate, organize,
and reach decisions in a realistic setting. For
example, one situational game is the supervisory
meeting in which each assessee sponsors a ficti-
tious candidate for a promotion. Since there is
only one promotion available, all assessees must
agree on the promotion of one individual. This
game is intended to test each assessee's ability

to lead, compete, and COOperate.

 

Typically six line managers one or more levels above

the individuals being assessed do the evaluating. In
some companies assessors get little more than an orienta-

tion on the assessment process. But most companies take

14
three or four days to train their assessors. The emphasis
Of assessor training is usually on observing behavior,
interviewing assessees, and conducting the in-basket
exercise. In addition, assessors usually practice on an
exercise given at the assessment center.

Seldom are professional psychologists used as
assessors, since,

The little research available indicates that
professionals do no better than trained line
managers in performing their tasks. While

the professional psychologists may have some
superior observational skills, this is probably
negated by their lack of company knowledge
(Byham, 1971, p. 13).

In most large companies, assessors serve only once.
However, smaller companies usually establish a pool of
trained assessors, drawing upon this pool for individual
assessors to serve as needed. One notable exception to

the general practice is in the Bell System where assessors

serve for six months.

Evaluation Processes

 

The data collected on each assessee are discussed at
a staff conference conducted at the end of the assessment
session. Assessors then rate the individual assessee on
several preestablished dimensions thought to be relevant

for managerial performance in the organization. In many

15
organizations, an overall rating of management potential
is also determined for each assessee. After all the
ratings have been decided upon by the staff, an assessment
report is prepared.

Practices with reSpect to assessment reports vary
from organization to organization. The most common prac-
tice is to have the assessment staff prepare a fairly—
detailed written narrative report on each assessee,
evaluating the individual's strengths and weaknesses for
future management positions. Emphasis sometimes is placed
on how to overcome any noted weaknesses, which enables the
report to be used as the basis for individual development.
Generally, the assessment data are made available to the
individual, as well as being retained for organization use.

One of the most important, yet most hazardous,
aspects of assessment center Operation is
feeding the reports back to the candidates.
Companies handle this in widely different
ways, depending on the purpose of their
centers (Byham, 1970, p. 158).

Some companies offer assessees the option of receiving
or not receiving feedback. Other companies provide feed-
back to all assessees as a part of their assessment pro-
cess. In some companies, assessees receive feedback prior

to departing the assessment center. In other organiza-

tions, they may wait some time for their feedback.

16
Typically, an assessor or a former assessor conducts the
feedback interview. However, if a staff psychologist is
available, in some organizations he has the responsibility

of discussing the assessment results with the assessee.

Research on Assessment Centers
There are several questions which need to be answered
before the value of assessment can be apprOpriately deter-
mined. Among these questions are:

1. To what extent is the assessment process a valid
predictor of management potential?

2. To what extent is the individual assessee
affected by the assessment process?

Validity of Assessment

with regard to the validity of assessment, there
appear to be a number of research reports which support
the assessment method of identifying management potential.
Undoubtedly the most extensive research is that conducted
by the American Telephone and Telegraph Company (AT&T) in
conjunction with its Management Progress Study. This study
was initiated in 1956 as a longitudinal study of the
develOpment of young men in a business environment. "Its
purpose is very general -- to learn more than is now known
about the characteristics and growth of men as they become,

or try to become, the middle and upper managers of a large

17
concern" (Bray, 1964, p. 420).

Subjects of this continuing study are 422 men, em—
ployed in six of the Bell System's telephone companies.
About two-thirds of the men are college graduates, who
were assessed soon after their employment. The other
third are men who started in the Bell System as vocational
employees, and advanced into management positions early in
their careers. Each subject spent three and one-half days
at an asSessment center, going through in groups of
twelve. The assessment center process was intended to
discover the abilities, aptitudes, motivational and per-
sonality characteristics, attitudes, and interpersonal
competence of potential managers. To determine these
factors, some 25 assessment variables were developed.
Techniques used to measure these variables were: inter-
view, in-basket exercise, small—business game, group dis-
cussions, questionnaires, projective tests, and paper-
and-pencil tests.

The assessment staff (usually nine professionally
trained individuals) assembled, reviewed, and discussed
the results. Typically, one to one and one-half hours
were devoted to evaluating each subject separately. The
subject was independently rated by each staff member on

the 25 variables. The staff also evaluated the man's

18
potential as a management person in the Bell System.
Since the data from assessment have not been used as
bases for promotion decisions, it seems reasonable to
argue that the relationship between assessment results
and promotions is not contaminated by the fact that the
decision maker has previous knowledge of assessment
ratings, and thus is using these data as a basis for pro-
motion rather than the traditional managerial evaluations.
There have been various research reports by AT&T

personnel concerning their Management Progress Study.
Bray and Grant (1966) reported on the management level
achieved and the current salary of the 422 subjects who
were still with the Bell System as of July 1965. Approxi-
mately one-fifth (21%) of the assessees had achieved
middle-management positions. In general, the college
group had progressed more rapidly than the non-college
group.

Of the 55 men achieving middle management,

43 (78%) were predicted correctly by the

assessors. In contrast, of the 73 men who

have not advanced beyond the first level

of management the assessment staffs pre-

dicted that 69 (95%) would not reach middle

management within 10 years (Bray & Grant,

1966, p. 18).

In a later report, Campbell and Bray (1967) examined

the subsequent job performance of men assessed five years

19

earlier. -They reported that 55% of the men promoted be—
fore the assessment program began were considered "above
average performers”, while 68% of the men who were assessed
as "acceptable" and later promoted to management were con-
sidered "above average performers".

This is a statistically significant finding

....It indicates that the assessment pro-

gram has been a definite aid in the selec-

tion of better performers at the first

level of management (Campbell & Bray, 1967,

pp. 10-11).

In yet another report from.AT&T, Bray and Campbell
(1968) described the application of the assessment center
method to the selection of prospective communication con-
sultants (salesmen). The assessment center evaluated
individuals who had been recently hired as salesmen. The
assessees had met all employment standards and had been
screened by their local Bell company as qualified for the
job of communications consultant. Assessment techniques
used were paper-and-pencil tests, an interview, and indi-
vidual/group simulations. Assessment staff judgments, as
to acceptability for sales employment, were used to place
each assessee into one of four categories: (1) more than
acceptable; (2) acceptable; (3) less than acceptable; and

(4) unacceptable.

Great care was taken that the results of the
man's performance at the assessment center

20
did not affect his assignment or appraisal
on the job. All of the men in the study
were Bell System employees at the time of
their assessment and there was no feed-

back to their trainers, their supervisors,

or to the men themselves on their perform-
ance at the assessment center (Bray &
Campbell, 1968, p. 37).

Assessment center judgments were compared against the
primary criterion of first-hand observation of actual
sales contacts made approximately six months after assess-
ment. The job performance observations were made by a
special Observational team working out of the AT&T head—
quarters in New York. A total of 78 assessees were evalu-
ated against review standards which included preparation,
usage prospecting, recommendations, closing, and imple-
mentation. As a result of the special Observations, each
man's performance was classified as either meeting stand-
ards or failing to meet standards.

Of the 9 men judged "more than acceptable" by the
assessment center staff, all met review standards. And
of the 21 men judged "unacceptable", only 2 passed the
field review. For the middle groups ("acceptable" and
"less than acceptable"), 26 of 48 assessees met the review
standards. These data indicated that assessment center
judgments were highly correlated with the field review

for the two extreme groups ("more than acceptable" and

21

"unacceptable"). Bray and Campbell reported that the
overall correlation between assessment center judgment and
subsequent field-performance ratings was .51.

These three research reports and others reported as
part of the AT&T management studies (Grant & Bray, 1968;
Grant, Katovsky, & Bray, 1967), indicate a relatively high
degree of predictive validity for assessment centers.
Other research studies provide corroboration of the AT&T
results. For example, at IBM, Wollowick and McNamara
(1969) reported a study where 94 lower and middle managers
were selected for assessment on the basis of having above-
average potential for advancement. Despite this restric-
tion of range, they found a correlation Of .37 with the
global assessment rating and the criterion of increase in
management responsibility three years after assessment.

Also at IBM, Kraut and Scott (1972) reported their
findings from a review of the career progress of 1,086
employees in sales, service, and administrative functions
of the Office Products Division of IBM. The subjects of
this review were employees who had participated in the IBM
assessment program from 1965 through the end of 1970. The
employees involved were from nonmanagement positions, who
were candidates for first—level management jobs. The re—

view was intended to evaluate the validity of the

22
assessment program. Two major criteria were used: (1)
second-level promotion; and (2) demotion from first-level
management.

Kraut and Scott found that, "among those who were
rated higher in the program, significantly greater propor-
tions go on to second positions" (Kraut & Scott, 1972,

p. 126).

Kraut and Scott also found that none of the sales em-
ployees in the highest assessment rating group were demoted,
compared with a 19% demotion rate in the lowest assessment
rating group. However, this difference was not statisti-
cally significant. On the basis of their review, Kraut
and Scott concluded:

The data collected thus far indicate this
large-scale assessment program appears
useful by making discriminations of man-
agement potential which are later confirmed
by the rate of promotions, as well as de-
motions (Kraut & Scott, 1972, p. 128).

Other organizations, such as Standard Oil of Ohio
(Carleton, 1970; Finkle and Jones, 1970), General Electric
(Meyer, 1970), and Sears Roebuck (Bentz, 1967) also have
established the validity of the assessment method for
identifying management potential.

However, there is one research report critical of the

assessment process. Hinrichs (1969) showed that ratings

23
of management potential based on a review, by two experi-
enced managers, of the personnel records of 47 IBM managers
correlated .46 with the ratings of management potential
these 47 managers had received after attending the IBM
assessment center. Based upon his finding, Hinrichs
(1969, p. 431) concluded,

The data suggest that traditional approaches
to the assessment of management potential in
the form of a careful evaluation of personnel
records and employment history ... can per-
haps provide much of the same information
which evolves from the lengthy and expensive
2-day assessment program.

However, Dunnette (1971, p. 106) disagreed with this
conclusion, and stated,

In my Opinion Hinrichs' argument, though
reasonable, cannot be sustained on the basis
of .46 he reports in his investigation.
Nearly 80 percent of the variance in the
assessment program rating remains unasso-
ciated with the ratings based on the per—
sonnel records; therefore, it seems highly
probable ... that the 'lengthy and expensive'
assessment program does contribute inde-
pendent, valid, and useful diagnostic infor-
mation about men's abilities and behavioral
tendencies that is not contributed by ratings
based merely on file information.

Impact of Assessment
Most research on assessment is aimed at the question

of validity of the technique. There is little research

which examines the impact of the assessment procedure on

 

II (III . Iain-ll. 5‘) \[III I
Ill] II ’0' .. I III

II 5’11" ll 4"

24
either the organization or the individual. One study
concerning the effects of an assessment center experience
is a research report prepared by the Pacific Telephone and
Telegraph Company. It was primarily concerned with deter-
mining if there were any long-range negative effects for
participants, attributable to their assessment experience.
In particular the researchers were interested in the
effects on those participants who had been rated "below
average" or "not acceptable".

A random sample of 99 men was selected for intensive
interviews by staff psychologists. Of this sample 47 were
”unsuccessfur'(had attended an assessment center but had
not been promoted) and 52 were "successful" (had been pro-
moted after attending an assessment center). The inter-
viewers used a focused interview technique to explore eight
specific areas for possible change in the assessee's life.

The most significant conclusion was that unsuccessful
assessees did find ways of adjusting to their poor perform-
ance in assessment. But the ways were not those expected.
Instead of becoming frustrated and giving up, the unsuc-
cessful assessees adjusted in more constructive ways, with
many of them (38%) appearing to use some form of rational-
ization. There was some evidence of long-range negative

effects on the unsuccessful assessees, in comparison with

25
the effects on the successful assessees; but on the whole,
positive effects seemed to predominate. Unsuccessful
participants appeared to have undergone a self-appraisal
which resulted in expanded involvement and increased self—
develOpment activity.

The research study concluded that, "there seems to be
very little reason for concern about any permanent damage
that assessment has caused with those who have not been
promoted" ("PAR Effects Study”, 1968, p. 30).

There are other research findings which tend to sup-
port this conclusion. For instance, Jaffee, Bender, and
Calvert (1970) indicated that participants in the Union
Carbide Company assessment program felt the program gave
an individual a fair chance to prove himself. And the
participants generally accepted the premise that poor per-
formance was the "fault" of the individual and not of the
system. Byham & Pentecost (1970) indicated that they have
found no conclusive evidence that candidates who do poorly

in assessment centers start looking around for another job.

Research Objective
The research evidence discussed in the last section
provide substantial evidence that, with varying degrees

of validity, prediction of useful criteria can be made

26

from multiple assessment information obtained at a per-
sonnel assessment center (Campbell, 1972). This is sup-
ported by Byham (1970), who concluded that the accumula-
tion of research findings from a variety of centers lent
considerable credibility to the overall validity of the
assessment technique. Byham (1970, p. 154) states:

In a survey of the 20 companies that operated

centers, I uncovered some 22 studies in all

that showed assessment more effective than

other approaches and only one that showed it

§§_effective as some other approaches. None

showed it less effective. As I suggested

before, these studies exhibit correlations

between center prediction and achievement

criteria such as advancement, salary grade,

and performance ratings that range as high

as .64.

However, some psychologists have questioned the pre—
dictive validity of assessment centers. Cronbach (1960)
claimed there is a definite problem in reconciling the
statistical evidence with the claimed "clinical" validity
of assessment techniques. Taft (1959) agreed that prob-
lems arise with respect to clinical versus statistical
predictions. He also asserted that problems arise from
conditional factors that affect the criteria.

Hardesty and Jones (1968) criticized most validity
methods because they are done after many personnel deci—

sions, based on the assessment information, have been made.

Bray and Grant (1966) supported this criticism, stating,

27
. . . where prior screening has been effec-
tive and/or the assessment results have
influenced personnel decisions allowance
for the consequent restrictions on range
of subsequent performance has been inade-
quate (Bray & Grant, 1966, p. 4).

Nevertheless, despite these criticisms, there are
strong supporters for the assessment center method of
evaluating management potential. Dunnette (1971, p. 16)
in his review of the literature on assessment centers
concluded that, "multiple assessment procedures for
identifying managerial talent have been shown to possess
the particular advantages suggested by their advocates".
Byham (1971, p. 16) agreed asserting that, "the assess-
ment center is a superior method of predicting management
potential —— compared with methods such as supervisor
appraisals and tests".

If the increasing use of assessment centers is an
indication, many organizations appear to also support the
assessment method of identifying management potential.
Perhaps a reason for this support, arguments about pre-
dictive validity notwithstanding, is that assessment has
face validity. That is, assessment provides,

. . . more 'real'measures of what a manager
might run into on the job than paper-and-
pencil inventories. Instead of measuring
traits or getting at behavior tendencies

indirectly, simulated procedures allow
direct observation of a man's behavior in

28
approaching what appears to be a highly
job-relevant but still fairly well-
structured and standardized stimulus
configuration (Campbell, et a1, 1970,
p. 142).

Assessment appears then to be an approach to the
identification of management potential which is more
readily understood and accepted,by both those undergoing
assessment and those using assessment results, than the
previous approaches which used a "traits" or "predictor”
concept to identify managerial talent. The attributes
of the assessment method (e.g., better prediction of
potential, assessee exposure to managerial demands and
reSponsibilities, identification of develOpment needs,
and assessor training), and the strong support assessment
centers have received from many industrial psychologists
and personnel managers, may account for the ever-
increasing use of the personnel assessment method for
identifying management potential.

However, despite this increasing use of assessment,
virtually no research has been directed at examining the
impact of such a process on the assessee. If the assess—
ment procedure does have an effect upon individuals who
are exposed to it, it may result in both direct and in—

direct effects on the organization. If it has negative

effects, turnover may increase. Assessment may heighten

29
expectations which may not be met in the future. This
could lead to frustration and dissatisfaction with the
organization.

The primary objective of the research is to examine
the relationship between the effects of assessment and
job performance and attitudes. Basically, the research
question is whether or not individuals,who are exposed to
assessment and get different kinds of feedback, react
differently in terms of the way they feel or the level
of their job performance. It is expected that this rela-
tionship might be affected by certain personality charac—
teristics. In addition, these characteristics may affect
the individual's job performance and job satisfaction.

A secondary objective of the research is to examine
the relationship between personality characteristics and
assessment center ratings. The reason for the interest
in this relationship stems from the fact that personality
characteristics or traits, as measured by personality in-
ventories, have been used in the past and are still being
used to some extent today to determine management poten-
tial. Since the assessment center is also used to deter—
mine management potential, some relationship might exist
between the two methods -- assessment and traits -— of

identifying management potential. Should such a

 

3O
relationship be found, then it is possible that results
of a personality inventory may be predictive of assessment
ratings. If this is so, management could substitute a
fairly inexpensive personality measurement for the more
costly assessment process, for purposes of identifying

management potential.

CHAPTER 2
RESEARCH DESIGN AND METHODOLOGY

In this chapter the underlying theoretical concepts
of the research are develOped. Then the model and hypothe-
ses generated by these concepts are presented. Finally,
the research design, research sample, measuring instru-
ments, and statistical methods employed in the research

are discussed.

Theoretical Concepts

As noted in the last chapter, with one exception no
research was found which dealt with effects on the
assessee resulting from participation at an assessment
center. Since the main purpose of this research is to
determine the effects on the assessee Of an assessment
center experience, it appears that the research will be
delving into relatively "unexplored" aspects of the
assessment center method of identifying management
potential. Accordingly, it is important to develOp some

concepts within which to conduct the research.

31

32

A convenient starting point is to consider the pur-
pose of a personnel assessment center. The primary pur—
pose Of the typical center, as explained previously, is
the identification of individuals with potential for
supervision. The underlying premise of this purpose is
that those who do well at an assessment center will move
into positions of greater supervisory responsibility.
Those who do poorly will remain in their present posi-
tion in the organization. Some may believe that those
who do poorly are "deadwood", and if they become discour-
aged and leave the organization, the organization bene-
fits. This latter position may not necessarily be sound
since it disregards the fact that the usual practice is
for supervisors to nominate individuals, who have shown
management potential through their job performance, to
attend an assessment center. Thus to the on-the-job
supervisor, the assessee who does poorly is certainly not
"deadwood"; rather he is a good worker who may lack super-
visory potential. Also the assessee may represent a size-
able investment in terms of experience or technical compe-
tence, and losing him may actually hurt the organization.

If the basic purpose of personnel assessment is to
identify those with management potential and not to eli-

minate those who do not exhibit this potential, it is

33
understandable that management concern over assessees'
reaction to an assessment center experience does exist
(Byham, 1971). Just what particular reaction may cause
concern is difficult to precisely discern because there
are many phases of the assessment method. And each of
these may cause varying reaction on the part of the
assessee. For example, the nomination to attend an
assessment center may well generate a sense of satisfac-
tion with the organization's procedure for selecting indi-
viduals for supervisory positions. In addition, the
physical facilities of the center, the measuring tech~
niques used, the extent of assessor qualification, and the
type of feedback interview may all bring about an assessee
reaction. Such reaction could be satisfaction or dissatis-
faction, or could be encouragement or discouragement with
the assessment procedures.

However, perhaps the single most important aspect of
assessment to both the assessee and the organization is
the final ratings given each individual by the assessment
staff. For this reason, this research will focus on the
effect of assessment ratings on reaction of individuals.
This seems to be reasonable since it has been found that,
"typically the greatest concern of management is the indi-

vidual who attends an assessment center and does poorly"

34
(Byham, 1971, p. 17).

This concern appears to be important because manage-
ment experience has suggested that those who do poorly,
with respect to some event which closely affects their
working career, often become discouraged about their future
with the organization. Such discouragement may generate
aftereffects impinging upon the assessee's interrelation-
ships with the organization. These aftereffects may vary,
but if they affect the individual's performance they are
likely to be a problem. This is based on the premise that
performance is the keystone of productivity; and in turn,
that productivity is the basic ingredient of an efficient
and effective organization. If assessment does have an
effect, the manager who sends someone to a center may
justifiably be concerned with the effects of assessment
on the subordinate's subsequent job performance.

In addition to effects on performance, the typical
manager is also concerned with the subordinate's job
satisfaction. Consequently, the manager may well ponder
the impact of assessment results both on his subordinate's
job performance and on his attitude toward the job, be—
cause the impact can alter the effectiveness of the entire

work unit.

These then are the problems with which the research

35
is concerned. When an individual undergoes a personnel
assessment what is the extent to which job performance and
job satisfaction change after he receives feedback about
assessment ratings? It is likely that those who do poorly
may experience negative effects. That is, they may be said
to have suffered adverse effects from the assessment eXpe-
rience.

However, it is unlikely that all individuals are
affected in the same fashion by feedback. The individual's
personality, especially his conception of himself,may be
an important factor which affects his reaction to assess-
ment. One approach to conceptualizing the notion of self-
concept is the idea of self-esteem.

One's self esteem refers to the extent to which the
individual perceives himself to be effective in dealing
with the problems that confront him (Ghiselli, 1955).

That is, one's self-esteem is the extent to which the indi—
vidual sees himself as a competent, need-satisfying indi—
vidual (Korman, 1970). So self-esteem is the value the
individual places on his image or concept of himself (Kay,
.EEJElr 1962).

Some individuals see themselves as being sound in
judgment and able to COpe with almost any situation

(Ghiselli, 1955). These are considered to have high

36

self-esteem (HSE). Other individuals think of themselves
as slow to grasp things, making many mistakes, and being
generally inept (Ghiselli, 1955). This group would be
considered as having low self-esteem (LSE). These indi—
vidual self-perceptions may be considered as relatively
persistent personality traits which occur relatively con—
sistently across various situations (Korman, 1970).

A person's self-esteem affects the evaluation

he places on his performance in a particular

situation and the manner in which he behaves

when in interaction with others. Self-esteem

concerns the amount of value an individual

attributes to various facets of his person

and may be said to be affected by the successes

and failures he has experienced in satisfying

central needs. It may be viewed as a function

of the coincidence between an individual's

aspirations and his achievement of these

aspirations. Self-esteem, then, may be

defined as the degree of correspondence

between an individual's ideal and actual

concepts of himself (Cohen, 1968, p. 383).

In general, HSE persons expect to be successful in
meeting their aspirations, while LSE persons expect to
encounter failure eXperiences. This implies that indi—
viduals with HSE might well react to new situations with
expectations Of success, since in the past they have been
successful in meeting their needs. And conversely, indi-
viduals with LSE might well react to new situations with

expectations of failure, since in the past they have been

unsuccessful in meeting their needs. Hence LSE's are more

37
vulnerable to the effects of failure experiences, which in
turn reinforces the general discrepancy between self-
ideals and self—percept (Cohen, 1968).

The levels of self—esteem may generate different kinds
of expectations. First, HSE persons may be less affected
by the communication of failure experiences and more
responsive to success experiences than are LSE persons,
since they may protect themselves from negative self-
evaluation and be less vulnerable to the impact of outside
events. Second, LSE persons may be more affected by what
others communicate to them concerning their performance,
since they are more apt to indulge in negative self-
evaluation and may be more vulnerable to the impact of
outside events (Cohen, 1968).

The level of self-esteem might also be important in
reactions to assessment. Korman has demonstrated that
self-esteem is a useful moderator to study other aspects
of organizational life. For example, his research has pro-
duced evidence showing that self-esteem is a moderating
variable in vocational choice (Korman, 1966), in task
success and task satisfaction (Korman, 1970). Also
Korman (1970) has hypothesized that the level of self-
esteem affects the satisfaction-performance relationship

in that at high levels of self-esteem performance predicts

38
satisfaction; whereas at low levels of self—esteem satis-

faction predicts performance.

Self—Esteem and Assessment

These considerations lead to some tentative hypotheses
concerning self—esteem as it may have an affect on an
individual's reaction to assessment ratings, and change in
job performance and job satisfaction after assessment.

First, considering those who attain "good" (above
median) assessment ratings, there should be different re-.
actions and aftereffects from the HSE person and from the
LSE person. For the HSE person, the ratings reinforce his
self-image because his expectation that he can successfully
overcome most situations in his work environment is con-
firmed. Thus his behavior patterns which led to his
ratings are reinforced, and consequently his performance
and satisfaction would be expected to rise as he continues
to follow what has been a successful path for him. For
the LSE person, positive assessment results indicate that
he has met with a "success". However, a success is basi-
cally not congruent with his self—image and he may tend
to disregard the assessment ratings, unless they are
reinforced later by other positive indications of his

behavior patterns (e.g., a promotion). SO consequently

39
his performance and satisfaction would be eXpected to
remain the same as before assessment.

Second, those who attain "poor" (below median)
assessment ratings, are likely to react differently as a
function of self-esteem. For the HSE person, low assess-
ment ratings are incongruent with his self-image and he
will tend to reject or "shake-off" this poor showing on
his part. While his self—perception of his behavior is
not reinforced in this instance, his performance and
satisfaction should tend to remain the same as before
assessment because he believes that the path he has
followed in the past has been generally successful for him
and will remain so in the future. For the LSE person, the
ratings confirm his self—belief that he is inept and un-
able to COpe with a new situation. He is now "proved un—
qualified" for increased management responsibilities by
the organization he works for. And since the LSE person
is vulnerable to failure, it can be expected that his per—
formance and satisfaction will decrease after assessment

as he follows that path he believes he fits.

Model and Hypotheses
Based upon the theoretical concepts develOped in the
last section, a model of the effects on an assessee of an

assessment center experience has been formulated (Fig. 2-1).

40

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M cofiuomwmfiumm K
mocmEuomumm I cmwpmz SOHOm
rxcofiuomwmwumm .k .\
n1 mocmBnOmumm A. cmfipmz m>on< '4 304
m cofluomwmfiumm
mocmauOmuom
ucmEmmomm<
..cowuumwmfiumm .\ agape: sonm 1T1. swam muomom
nllloocmEDOuuom .4. . ammummimamm
cOHuOmwwHumm
e mocmEMOMAOm AT cmwpmz m>on<
+

 

 

ucmEmmwmm< umum<

 

 

 

lwdwumm ucmEmwomm¢

mmmmmmm< z< 20m: meummmm HszmmMmm< mo AMQQZ

Hum MMDOHM

41

In the model, a moderating variable -- self-esteem --
is hypothesized to affect the nature of the reaction to
assessment ratings. Essentially it is hypothesized that
where assessment ratings positively reinforce the
assessee's self—esteem, job performance and job satisfac-
tion change after assessment; and where assessment ratings
are incongruent with or negatively reinforce the assessee's
self-esteem, job performance and job satisfaction remain
unchanged after assessment.

More specifically, it is hypothesized:

1. For the HSE person who receives above median
assessment ratings, job performance and job
satisfaction will increase or change in a
positive direction after assessment.

2. For the HSE person who receives below median
assessment ratings, job performance and jOb
satisfaction will not change after assessment.

3. For the LSE person who receives above median
assessment ratings, jOb performance and job
satisfaction will not change after assessment.

4. For the LSE person who receives below median
assessment ratings, job performance and job
satisfaction will decrease or change in a

negative direction after assessment.

42

The hypotheses are based on two assumptions:

1. Job satisfaction is in some way related meaning-
fully to job performance. This assumption is
supported by Porter and Lawler (1968) who took
the position that there is a relationship between
job performance and job attitudes, which they
assert is not necessarily a causal relationship
but is a consistency of direction relationship.

2. Before assessment jOb performance levels for
all assessees are satisfactory or better. Hence,
if the assessee maintains this level of perform-
ance after assessment, he may still be regarded
as a "good performer". The assessee is not
considered to have experienced an adverse effect,
or to have "failed", if his job performance level
does not increase after assessment.

Research Design

The hypotheses presented in the last section basically
assert that an assessee's jOb performance and jOb satisfac-
tion will be affected by assessment center ratings and the
assessee's level of self—esteem. Expressed in other words,
it is hypothesized that an assessee's level of performance
and satisfaction may be different or may remain unchanged
after assessment, when compared with his before assessment
levels of performance and satisfaction. So in order to
test the hypotheses, it is necessary to compare assessee
performance and satisfaction at two points in time:
(1) before the assessee knows his assessment center

ratings; and (2) after the assessment center ratings have

been given to the assessee.

43

Primary Research Objective

The research design for testing the hypotheses re-

quires a "before-after" measure of assessee job performance

and job satisfaction. Any changes in performance and

satisfaction then can be analyzed to determine the extent

to which these changes fit the pattern predicted by the

hypotheses.

Because it is hypothesized that changes in job per-

formance and job satisfaction will vary according to

assessees' assessment rating and level of self-esteem, it

is necessary to divide assessees into groups according to

these two factors. This grouping of subjects was accom-

plished by first dividing subjects according to self-

esteem and then according to assessment results. In both

cases the median score or rating was used as the dividing

point.

1.

Self-esteem: The median score on the self-
assurance scale of the Ghiselli SDI (used to
determine self-esteem)was 30. This score is

the 56th percentile rank according to Ghiselli's
norms (Ghiselli, 1971, p. 145). The next lower
score -- 29 -- is the 47th percentile rank. So
the median used here approximates that of the

population from which Ghiselli derived his norms.

44
Subjects falling above the median score were
placed in the above median group, and were desig-
nated high self-esteem (HSE). The subjects
falling below the median score were placed in
the below median group, and were designated
.1ow self-esteem (LSE).

2. Assessment results: The median overall assess-
ment rating for the assessees was 5.25. Those
falling above the median were considered as
demonstrating above norm (+) management potential,
while those falling below the median were con-
sidered as demonstrating below norm (-) manage-
ment potential. Both the HSE and LSE groups
were further divided according to assessment
rating. At this point four groups of subjects
had been identified and designated as indicated

in Figure 2-2.

FIGURE 2-2

IDENTIFICATION AND DESIGNATION OF ASSESSEE GROUPS

 

 

 

Desig-
Group nation
1. high self—esteem, high assessment rating HSE+
2. high self-esteem, low assessment rating HSE-
3. low self-esteem, high assessment rating LSE+

4. low self-esteem, low assessment rating LSE-

 

45

Since it was hypothesized that assessees would differ
systematically as to assessment ratings and self-esteem
over time, a three-way analysis of variance was used to
test the hypotheses. That is, a 2 x 2 x 2 factorial design
was employed. The main effects were assessment center
rating (high: +, low: -), self-esteem (high: HSE, low: LSE),
and time (before assessment, after assessment). The
dependent variables were job performance and job satis-
faction.

The design may be portrayed as shown in Figure 2-3.

HSE

Self— .
Esteem

LSE

46

FIGURE 2-3

SCHEMATIC OF ANALYSIS OF VARIANCE

 

 

 

 

 

 

 

 

Job Job
Performance Performance
Job Job
Satisfaction Satisfaction
Job Job
Performance Performance
Job Job
Satisfaction Satisfaction
Before After
Assessment Assessment

Time

_._._

 

High (+)

 

Assessment

Low (—)

47
Secondary Research Objective

The basic research design examines the primary Objec-
tive of the research: determining the effects of assess-
ment ratings and self-esteem upon assessee job performance
and job satisfaction before and after assessment feedback.
The relationships among personality characteristics and
various measures of job performance and job satisfaction
were also considered. This phase of the research was con—
ducted using: (1) correlational analyses of personality
traits and various job performance measures; and (2) cor-
relational analyses Of personality traits and various job
satisfaction measures.

The second Objective Of the research was to determine
if measures of personality characteristics could poten-
tially be substituted for an assessment center evaluation,
as a method for identifying management potential. Such an
Objective required examination of the relationships among
personality traits and assessment ratings. Accordingly,
the research design for this phase of the research called
for correlational analyses of personality characteristics
and assessment ratings rendered by the assessment center

staff.

48

Research Sample

 

Subjects for the research were 60 males employed by
the parts division of a large manufacturing firm in the
automotive industry. On the average these men were 35
years old, had completed college, had been with the organ-
ization nine years, and earned $20,000 a year. Twenty-
four of the men worked in district sales offices located
throughout the United States. They represented such func-
tional areas as customer service, marketing, advertising,
merchandising, and sales. All could be considered as
potential district sales managers. The other 36 men
worked in parts depots also located throughout the United
States. They represented such functional areas as traffic,
warehouse Operations, packaging, Operations planning, and
systems coordination. All of these men could be considered
as potential depot managers.

Subject shrinkage was experienced during the research
because some of the before assessment measures were not
completed by all subjects, and because some supervisors
and subjects did not complete the after assessment mea-
sures. For the analysis of variance portion of the
research design, Table 2-1 indicates the N's of the four
groups of assessees for the measurement of the dependent

variables.

49

TABLE 2-1

SUBJECTS INVOLVED IN ANALYSIS OF VARIANCE
OF VARIOUS EFFECTS OF ASSESSMENT

 

 

Dependent Variables

 

 

 

*
GIOUP Job performance Job satisfaction
(N) (N)
HSE+ 11 ll
HSE- 6 7
LSE+ 10 5
LSE- 12 .9.
Total 42 32
* HSE+ : high self—esteem, high assessment rating
HSE- . high self-esteem, low assessment rating
LSE+ ' low self-esteem, high assessment rating
LSE- : low self-esteem, low assessment rating

For the correlational analysis phase of the research

design, Table 2-2 indicates the N's involved in the

examination of the various measures.

TABLE 2—2

SUBJECTS INVOLVED IN ANALYSES OF GHISELLI

 

 

 

CHARACTERISTICS AND VARIOUS CRITERIA
Measure N
Assessment Evaluation 56
Job Performance 56
Job Satisfaction 34

 

50
Assessment Program

The division's assessment program is much like that
"typical" assessment center described in the first chapter.
However, it should be noted that the primary objective of
the assessment program was to identify managerial strengths
and weaknesses for the purpose of planning develOpmental,
training, or experience needs in order to build a more
effective managerial force in the organization. The center
was thus oriented toward the identification of specific
develOpmental needs in order to enhance the organization's
pool of managerial talent, as Opposed to the mere identifi-
cation Of individual managerial potential as is the case
for most centers. Nevertheless, the personnel staff of
the division recognized that information generated at the
center would be a part of the decision process relating to
promotions. The assessees themselves regarded center
results as important to their future careers in the
organization.

Other than that, the center was Operated much like
other assessment centers. Assessees were selected on a
random basis to attend an assessment session. The
assessees, who went through the center in groups Of twelve,
began their experience at the center with a group discus-

sion meeting, during which the center administrator

51
outlined in detail the purpose and conduct of the center
and the dimensions or variables to be measured. Also dis-
cussed at this meeting were the roles of participants and
assessors.

Assessors were six managers or supervisors, usually
two levels removed from the assessees and not related to
the assessees in a direct supervisor-subordinate relation-
ship. The assessors received a week of intensive training
to develOp skills of evaluation. They usually served as
evaluators for one assessment session.

Assessees were evaluated over a period of two days on
ten variables. Assessment techniques used were: (1) in-
basket; (2) problem solving; (3) leaderless group discus-
sion cases; and (4) two leaderless group discussion
assigned roles. Figure 2-4 indicates the specific vari-
ables evaluated, the techniques used for evaluating each
variable, and the assessee behavior the assessors were

instructed to look for during assessment.

52

wmaouucoo

m>wumuuwficfi8pm nwwabmumo pcm .muficsu
nuoaao ocu cm>Hm cmn3 .mummoaop on moool
wmuwcuo mo xuoa gnu mNficmqu can coda
Ou zufiafibm mzu 302m Hmspfi>wpcfi OLD moon:

NOOHmfiomp

OLD mwo mcwuusd cmnu umzumu .wabmafim>m
coaumsDOLcH Ono mo mammb one so pcOQOOH
pcm samumwpossfi mummmwooc OH cOHmHomp
m cogs Onwcwoomu Hespfi>fipcfi mnu moon:

mmuomw ucm>maouuw pcm ucm>onu
cmozuwb mcaumcfiafiuomap mo mabmdmo o: mHI
wmeuHuOHud Hamwcficmwa Oucfi
xu03 we: mo mucmawam use ou mans on mHI
wmucoawpsfi pcsom
oxma cam .muumm ucmcwuumd mumsam>m
6cm boo xmmm Ou mans Hmspw>wpcfi onu mHI

wmmon m Oman uoc
.umpmwa m up Be: Ou pcOdmmu OHdOmd on:
chHuooqu now EH: 0u xooa manomd on:
mxuflawumon wcHumwuo usonuHB

xmmu O NO cowusHOm one ca EH: SOHHOm

Ou deOOd bow Ou mabm Hmspw>fipcfi mnu mHI

Suwafin< HmCOfimeHcmwuo

mwmco>fimfiomn

wcwxmz cowmfiowo

aficmAOpmog

 

Mom poxooq
pow>msmm mommmmm<

nus mace nos wea>som
pmcwfimmm mmmmo Emaboum

uoxmmm
ICH

 

manmwum> mumsam>m
ou pom: Ongflccuma

oanmfium>

 

 

mmezmo HZmmemmm< mmH mo mMDOHszmH Qz< mmqm<Hm<>

«IN MMDUHM

53

mmuoxuoauoo mo mmmc
Im>fiuomwwo mcu a: Ouwm Ou maps on mHI
wuomumucw owns on 8053 :ua3 muocuo uo
mommocxmo3 cam mnuwcmuum mnu w>fiwuumd
zamumusoom Ou mans Hmspfi>fipcfi ecu mHI

mxflaammmooosm cowumsufim Ono cues

Hmop Ou mem on ma no .muoooo mwcmcu
coppsm m coca possumsam osooon o: moon:
emcowuooufip

no mcowuom O>Humcuouam mumuocmw Ou

OHbm Hespfi>wpcfi ecu ma .OOAumsuowcH 3w:
no mcoauwpcoo mcfiwcmno nufis poomm cons:

wwcflaaou Hana mnu bow Ou wco msu on mHI
wumuumumlmamm m on mHI

masouw OLD cufi3 wcoam ow on

moop no muco>o no manomd mucosamcw ob
unaduum >Ho>wuom Hespw>apcfl ecu moon:

wmcmad was NO muumwwm owcmu
chH OLD ucsouom OucH oxmu on moon:
News on cmo mmcwa
upmmc umnu Om mmaspmzom a: bow on mmoal

ucwscnmowfia

OOOfiqucoo
wcfiwcmsu Ou omdoammm

w>aumauacH

 

pom poxooq
uoa>mzmm mmmwwmm<

nos «Hoe nus maa>aom
pmcwwmm< mommo SOHboum

umxmmm
ICH

 

oanmfium> oumsflm>m
ou pom: mogwcnOmH

OHbmfium>

 

 

 

pmoaflucoollqlm mMDUHm

54

 

«consumes umaamuw was mHI
wmucfloa AOHOB unwfianwfin on moon:
NEAOM wanmpcmuw

lumps: paw Hmuwwoa m ca cOfiumSuoucH

 

souufius ucomoud Hmspfi>apcw ecu cmuu x x coaumOHGOEEOU swuuHHS
NpONHcmwuo
Ham3 paw ucmpwwcoo pcsom o; moon:
we“: nufia cowummum>
Icoo m cfl pmwmwcw coca DOObm wcaxamu
ma on umna pOOOOMOOcs madooa on:
wcowumsuww wcfiumoa
m ca 3mH> mo ucfiOQ was ucommud AHO>HD
sommmo pcm xaumwao HmOpH>Hch one cmol x x x x coaumuficsaaou Hmuo
Qua mace nos wea>aom Sesame
pom poxooq pmcwwmm< mommo Emaboum IcH

pofl>mcmm mommwmm<

 

mabmfium> Oumaam>m
Ou pom: osvflcnooy

managem>

 

 

pwscfiucOUIlclm mmonm

55

During the second two days of an assessment session,
the assessors evaluated the assessees' assessment center
performance. Each assessee was rated, on the ten vari-
ables evaluated, on a nine-point scale. No overall rating
was given. subjectively, an assessee was considered to
have attained "good" results if he received a rating above
6 for a measurement factor. He was considered to have
attained "poor" results if he received a rating of 3 or
below for a measurement factor. (See Figure 2-4 for list
of factors.) Those receiving ratings in between were con-
sidered "average".

A written report, summarizing the findings and
Opinions of the assessors, and containing develOpmental
suggestions and recommendations, was prepared for each
assessee. Only an original of the report was made, and
it was retained by the personnel planning office of the
division as a permanent part of the firm's management
develOpment file.

The results of the center were given to the assessee
by a member of the center staff about two weeks after an
assessment session. Each assessee was given a list of
the ten variables evaluated and the assessment center
staff's ratings on these variables. The assessee was

encouraged to take notes, but he was not given a COpy of

56

the final report. In keeping with the primary purpose of
the assessment center, emphasis during feedback was placed
on overcoming weak areas noted by the assessment staff.
Stress was placed on how a develOpment program could in-
crease the assessee's on—the-job effectiveness, which in
turn would enhance his managerial capability and potential.
The approach used in feedback appeared to be that it was
better to detect and to correct weaknesses now, than to
be embarrassed and possibly hindered by them later.

No report was given to the assessee's supervisor,
other than an oral summary of the assessment ratings and

an outline of a recommended develOpment program.

Measuring Instruments
Various measures were used to determine: (1) per-
sonality characteristics; (2) assessment performance;

(3) job satisfaction; and (4) jOb performance.

Personality Characteristics

 

The Ghiselli Self-Description Inventory (Appendix A)
was used to determine the individual assessee's personality
characteristics. The Self-Description Index (SDI) was
administered during an assessment session before assess-
ment results were known.

For determining the effects of assessment on the

 

57
assessee job performance and jOb satisfaction the SDI self-
assurancesxsle was used, as it was by Korman (1970), to
measure the individual assessee's level of self-esteem.
For the secondary objectives of examining the relationships
among personality characteristics,job performance ratings,
job satisfaction scores, and assessment ratings, the SDI
provided a personality inventory which has been frequently
used by other researchers.

The SDI, according to Ghiselli (1971), measures indi-
vidual abilities, traits, and motivations found in the
successful manager. As defined by Ghiselli (1971), these
are:

I. Abilities:

l. Supervisory ability: capacity to direct the
work Of others, and to organize and inte-
grate their activities so that the goal of
the work group can be attained. '

2. Intelligence: cognitive capacity of the
mind involving such capacities as judgment
and reasoning; and the capacity to deal
with ideas, abstractions, and concepts.

3. Initiative: has two aspects: (a) ability to
act independently and ability to initiate
actions without stimulation and support from
others; (b) capacity to see courses of action

and implementations that are not readily
apparent to others.

 

II. Personality Traits:

4. Self—assurance: extent to which the indi-
vidual perceives himself to be effective

58

in dealing with the problems that confront
him.

5. Decisiveness: extent to which an individual
sees that a decision must be made and goes
ahead and makes it.

6. Masculinityefemininity: extent to which an
individual of one sex manifests the traits,
perceptions, or other qualities associated
with members of the Opposite sex.

7. Maturity: that state where the processes of
develOpment are complete so that there is no
further natural growth or improvement.

8. Working class affinity: extent to which the
individual is likely to be accepted or re-
jected by those of the working class as a
suitable person to associate with.

III. Motivations

9. Need for occupational achievement: desire to
achieve the responsibility and the prestige
which is associated with high position. (This
trait is sometimes referred to as achievement
motivation.)

10. Need for self-actualization: desire to
utilize one's talents to the fullest extent.

11. Need forypower: desire to direct and control
the activities of others.

12. Need for high financial reward: desire for
monetary gain from one's work.

13. Need foryjob securipy: extent to which an
individual is fearful of his circumstances
and wants protection from adverse forces.

Ghiselli (1971) has reported validity coefficients

between scores and job successes as shown in Table 2-3.

59

TABLE 2-3

COEFFICIENTS OF CORRELATION BETWEEN THE SCORES
0F MANAGERS, SUPERVISORS, AND WORKERS ON THE
VARIOUS SDI SCALES AND THEIR JOB SUCCESS

 

Managers Supervisors Workers

 

Supervisory ability .46 .34 .10
Intelligence .27 .06 .03
Initiative .15 -.O7 .02
Self-assurance .19 .18 -.03
Decisiveness .22 .15 .05
Masculinity—femininity -.05 -.O7 -.O9
Maturity -.03 .13 .02
Working class affinity —.17 .07 -.03
Need for occupational achievement .34 .08 .01
Need for self-actualization .26 -.03 .05
Need for power over others .03 .12 -.16
Need for high financial reward —.18 -.05 -.10
Need for job security -.30 -.05 -.ll

 

Source: Edwin E. Ghiselli, Explorations in Manggerial Talent
(Pacific Palisades, California: Goodyear Publishing,
1971), p. 150.

In developing the data shown in Table 2-3, Ghiselli
used 306 managers, 111 line supervisors, and 238 line
workers drawn from a wide assortment of business and
industrial firms located in various parts of the United
States. These individuals were administered the SDI and
also rated by their superiors. The SDI scores were cor-
related with judgments of the superiors, which had been
divided into two categories: (1) more successful; and

(2) less successful (Table 2-3).

60

Assessment Performance in the Present Study

 

The assessment center ratings given each assessee in
the present study by the cooperating organization's assess-
ment staff were used as a measure of assessment perform-
ance. The staff evaluated assessees on variables or
dimensions which in general included the following:

The ability to plan, organize and control effec-
tively.

The ability to work with others and to influence
them to a course Of action.

The ability to exercise leadership in a group --
or to contribute materially to a group's goals.

The capacity to learn from, or to use written
and oral communications.

The ability to adjust to changing conditions.

Sensitivity toward the Opinions and feelings of
others.

The ability to analyze data, solve problems or
arrive at decisions.

The specific variables evaluated are shown in Figure
2-4.

As previously indicated, each assessee was rated on
all ten variables by the assessment staff, with each
rating made on a nine-point scale. The points on the
scale were defined for the assessors as follows:

1 -- Low: Shows considerable negative behavior

in this particular skill area or has

consistently not dis la ed expected .
behav1or when the $1 ua ion required It.

61

3 —- Below Average: Shows little of this
skill and would definitely need
development in this area.

U)
I
I

Satisfactory: Displays an adequate
amount of the skill but could prob-
ably use some develOpment in the
area; but there is nothing to indi-
cate that he has difficulties in
the skill.

7 -- Above Average: Displays particular skill
to a greater degree than many
presently functioning managers.

The skill is displayed strongly.

9 -- Exceptional: Shows as much of the skill
as could be expected. Very well
prepared for a management job con-
sidering only the particular skill
in question.

An inter-item analysis, using Pearson's moment cor-

relation,was made of these assessment ratings (Table 2-4).

62

 

 

 

 

 

oo.a as. am. um. 08. mm. me. am. as. no. es nonsmoaesaaoo amuse“:
oo.a we. no. me. am. as. me. so. me. a coaumoaasaaoo Hmuo
oo.H me. co. co. Se. so. mm. Ne. w namaeumuman
oo.a am. as. as. am. me. am. A menace on mmeoamom
oo.a we. mm. mm. ok. km. S m>aumauaea
oo.H mm. Ne. He. SA. A coaumwmama
oo.H ea. om. mm. s kneaana Hmeoaumuaamwuo
oo.H as. as. m mmmam>amaumo
oo.H an. N messes eoamauma
oo.a H aaemnoemma
OH a w a e m a m N a smuH
BwuH
Aemuzv

monH<Ammmoo ZmBHlmmHzH mq<um UZHH<M Hzmzmmmmm<

«IN mqm<B

 

63

Due to the extremely high inter-correlations among
the ten factors, it was decided that they should be com-
bined into one scale. The internal reliability of the
combined assessment rating scale was computed (Nunnally,
1967, p. 193) to be .96. This high level of internal
reliability suggests that it is apprOpriate to use the
combined index as an assessment rating, rather than the

ten individual ratings.

Job Satisfaction

The Job Descriptive Index (JDI) (Appendix B) develOped
by Smith, Kendall, and Hulin (1969) was used to determine
the assessee's attitude toward such job related factors
as work, supervision, promotions, and co-workers. The JDI
was administered during an assessment session, providing a
before assessment measure of job satisfaction. A second
JDI was administered 180 days on the average after assess—
ment, providing an after assessment measure of job satis-
faction.

For the four JDI scales used, Smith, et a1 (1969)

report reliabilities and validities as shown in Table 2-5.

64

TABLE 2-5

RELIABILITIES AND VALIDITIES
0F JDI SCALES

 

 

 

 

Scale Reliability Validity
Work .84 .75
Supervision .87 .78
Promotions .86 .64
Co-Workers .88 .57
Source: Patricia Cain Smith, Lorne M. Kendall, and

Charles L. Hulin, The Measurement of Satis-
faction in Work and Retirement (Chicago:
Rand McNally, 1969), p. 74 and pp. 87-147.

Job Performance

 

Performance rating forms develOped (for purposes of

evaluating their assessment program) by the personnel re-

search staff Of the organization COOperating in the

research were

1.

used. Three different forms were employed:

Self-Performance Rating Form (Appendix C) --
completed by assessee during an assessment
session before assessment results were
known.

Performance Rating Form (Appendix D) --
completed by assessee's supervisor and

two co-workers before assessment. (For
purposes of this research, the two co-
workers' ratings were averaged and are
subsequently referred to as "peer" ratings
or "peer-average" ratings.)

Performance Rating Form (Appendix E) --
similar to second form —- completed by

65

assessee's supervisor 180 days on the
average after assessment.

Only supervisor ratings were obtained both before and
after assessment. Consequently, it was necessary to use
these ratings to measure assessee job performance. The
other performance ratings were used to evaluate the
validity of the supervisor's rating.

The ten factors or variables listed on the three per-
formance rating forms are the same as the variables evalu-
ated by the assessment center staff. These factors,
previously described and discussed in the section on the

research sample, are:

Leadership

Decision Making

Decisiveness

Organizational Ability
Delegation

Initiative

Response to Changing Conditions
Discernment

Oral Communication

Written Communication

OkOCDflOWU'l-PWMl-l

f—J

An inter-item correlation analysis, again using
Pearson's moment correlation, was performed on the three
measures (self, supervisor, peer) of before assessment job
performance. These correlations are shown in Tables 2-6,

2-7, and 2-8.

66

a
.«mxauxv + H

 

 

 

.Amaa .a .keasv asamaasz . u axe
hams

"masauow he peasdsoo cm. H huwawnmwamm assumucH
oo.H as. as. mm. em. mm. me. mm. om. mm. 0H cowumuaesesou emanate
oo.a mm. mm. as. om. He. om. am. am. a coaumuaaSBEOO Hmuo
oo.H on. em. as. we. om. em. as. m nemaeumomeo
oo.H SN. NN. mm. em. mm. mm. a manage on mmaoammm
oo.H om. mm. mm. me. me. e m>aumauaaH
oo.H am. am. SH. me. n eowumwmamn
oo.H Na. Na. as. e knaauna Haeoaumuaamweo
oo.H me. 00. m mmmew>amauma
oo.H as. N messes cowmaumn
oo.a H aaemuoemma

0H m m k e m a m N H

 

Em
EmuH OH

 

 

Aomuz I undammmmm< ouomwmv
monH<qmmmou ZmHHummth mq<um ozHH<m moz<zmommmm ozHB<mumqmm

elm m4m<H

67

a
.mesnsv + a as

 

 

 

 

.Amas .a .Nomsv sssaccaz . man u t
.x

”mflsshow mp pmusdaoo em. u SDNHNQONHOM accumucH
oo.a mm. NN. SN. as. ON. me. as. oe. Nm. OH coaumueesaaoo ambush:
oo.a NN. so. NN. NN. NN. as. NH. me. a coaumuaessaoo Hmto
oo.N mN. em. on. mm. mo. as. mm. m unmannmomNa
oo.N NN. AN. ON. me. om. NN. N manage on mmcoammm
oo.N NN. em. as. as. me. S m>aumNuHeH
oo.H oe. ma. mo. mm. m acaumwmsma
oo.a NH. Nm. we. a auNaNna Nmeoeumuaemwuo
oo.N mm. om. N mmmem>umaume
oo.N ms. N Nessa: aoamwome
oo.H a aaemnmemma

ON a N N e m e N N a amuH

EOOH

Aomnz I undammoww< ouomomv
monH<Ammmou SMHHIMmHzH mq<om UZHH<¢ moz<zmommmm MOmH>mmmbm

mIN mqm<H

68

N
.NwNNINS + N xx

 

 

 

 

.ASSN .a .NSSNS NNchcsz . SNN u S
IS

"mHOENOw Sb pmusdEOu mm. H SNNHNQONHOM HchODCH
oo.N SS. SN. SS. SS. NN. NS. NN. SN. SS. SN SONSSoNasaaou SSSSNNS
oo.N SN. SS. Ne. NN. SN. SS. SS. SS. S SONSSUNaaaeoo Nmeo
oo.N SS. SN. NS. SN. SS. NN. om. S SemactmumNa
oo.N NS. SN. Se. SS. SS. as. N SSSSSS on mmaoamum
oo.N SN. NN. SS. SN. NS. S S>NSSNSNeN
oo.N NN. SS. SN. NN. S :ONSSSSNSS
oo.N SS. SN. SS. S NSNNNSS NmaoNSSSNcSSNo
oo.N SS. Ne. S SSSSS>NSNSSS
oo.N SS. N SchSz eoNSNuma
oo.N N aNSSSSSSmN

SN S S N S S e S N N aSSN

EmuH

 

 

Aomuz I ucmammmwm< ouowmmv

monH<4mmmoo ZmHHImmHzH m4<um OZHH<M muz<zmommmm mu<¢m>€lmmmm

wIN mude

69
Since the internal reliabilities of all three scales
were high, it was decided that one overall rating would
be used for each of the scales. This overall rating was
the average of the ten individual ratings on each scale.
As explained before,

self-ratings and peer-average

ratings were used to help estimate the validity of the

supervisor's judgment of assessee job performance. To
make this estimate, intercorrelations of the three
scales were computed, and are shown in Table 2-9. Also

shown in Table 2—9, are the intercorrelations of the

overall assessment center rating with the three before

assessment overall job performance ratings.

TABLE 2-9

INTERCORRELATIONS AMONG ASSESSMENT RATING
AND VARIOUS JOB PERFORMANCE RATINGS
(Before Assessment - N = 56)

 

 

 

 

Rating Assessment Peer-Average Supervisor Self
Assessment 1.00

Peer—Average .29* 1.00

Supervisor .30* .52** 1.00

Self .44** .08 .30* 1.00
* Significant at .05 level

** Significant at .01 level

70
Table 2-9 indicates a relatively high degree of agree-

ment exists between supervisors and peers, and a moderate
degree of agreement exists between supervisors and self-
ratings, as to assessee job performance. While peer and
self-ratings were not correlated, the pattern of multi-
rater agreement suggests that the use of the supervisor's
performance rating is an acceptable measure Of before

assessment and after assessment job performance.

Statistical Methods
Three statistical analyses used for the research
were: (1) analysis of variance; (2) Scheffe's post hoc
method for comparison of means; and (3) correlational

analysis.

Analysis of Variance

Analysis of variance was used because such analysis
provides a method for the simultaneous comparison of many
means in order to determine if some statistical relation
exists among the variables involved. A major advantage
of this statistical method is that reasonable departures
from the statistical assumptions of normality and homoge-
neity will not seriously affect the validity of the

inferences drawn from the data.

71

Scheffe Test

 

After an analysis of variance, when a significant F
ratio is present, interpretation of the data Often requires
a comparison of pairs of means. The differences between
some pairs may be significant, while the differences be—
tween others may not be significant. The research design
used in this research calls for this type of data interpre-
tation. Accordingly, the Scheffe method of means compari-
son was used when a significant F ratio was found.

The Scheffe method was selected because no special
problems arise from unequal N's, and results are not
seriously affected by moderate violations of the assump-
tions of normality and homogeneity of variance. Finally,
the method uses more rigorous standards than other multiple
comparison methods, so fewer significant differences re-
sult. Because it is more rigorous than other procedures,
when using the Scheffe method the common practice is to
use the .10 level of significance instead of the .05 level
(Ferguson, 1966, p. 297). This practice was followed in
analyzing the present research data.

All combinations of means were subjected to the
Scheffe test; however, only those means with significant

differences are reported.

 

72

Correlational Analysis

 

Correlation coefficients were obtained using Pearson's
product-moment correlation. The degree of relationship is
provided by the product-moment correlation coefficient
"r", which is the average product of first moments of two
distributions. The sign and size of r, which cannot be
greater than +1.00 or less than -l.00, provide a very
understandable indication of the direction and degree of
relationship between two variables (Nunnally, 1967).

Correlational analyses were used to determine the
inter-item correlations among the various performance
scales, and to determine the extent of relationship among
these scales. In addition, correlational analyses were
used to determine the extent of the relationship among
personality traits and assessment ratings, among person-
ality traits and job performance measures, and among

personality traits and job satisfaction measures.

Summary

In this chapter, the underlying theoretical concepts
of the research, and the model and hypotheses generated
by the concepts, were presented. The research design, the
research sample, the measuring instrument used, and the

various statistical methods employed in analyzing the

73
research data, were discussed. Now to be considered are
the results of the research, which are presented in the

next chapter.

CHAPTER 3

RESEARCH RESULTS

This research has two objectives. The primary Objec—
tive is to determine the extent to which an assessment
center experience affects assessee job performance and
job satisfaction. In Chapter 2, certain effects on indi-
viduals were hypothesized, and these were expected to be
different as a function Of the assessee's level of self-
esteem and level of assessment rating. To test the
hypotheses, a three-way analysis of variance was performed.
The independent variables were self—esteem, assessment
ratings, and time. The dependent variables were job per-
formance and job satisfaction. This chapter reports the
results of the analyses of variance.

This chapter also reports results of the examination
of the relationships among characteristics measured by the
Ghiselli SDI and job performance and job satisfaction.
Additionally, Ghiselli SDI dimensions were correlated with
assessment ratings. Since these latter analyses were
exploratory in nature, no specific hypotheses were

74

75

develOped.

More extensive discussion and interpretation of the
research results follow in Chapter 4.

The Impact of Assessment on Performance
and Satisfaction

In this section, a series of five tables presents the
results of the analyses Of variance concerning assessment,
self-esteem, and time effects on assessee job performance
and job satisfaction. Also shown in each table are the
means for the four groups of assessees. Where a signifi-
cant F ratio occurs, the Scheffe comparison of means is
used to determine which difference between all combinations

of means is significant.

Impact of Assessment on Job Performance

Table 3-1 reports the ANOVA results with respect to
supervisor job performance ratings, examining the relation-
ship between the dependent variable supervisor performance
rating and the main effects of assessment rating, self-
esteem, and before-after. For this analysis, performance
data were available for 42 subjects who were rated both
before assessment and after assessment by their super-

visors.

RELATIONSHIP BETWEEN:
VISOR JOB PERFORMANCE RATING, AND (C) ASSESSMENT RATING FOR

76

TABLE 3—1

(A) BEFORE AND AFTER ASSESSMENT, (B) SUPER-

ASSESSEES AT DIFFERENT LEVELS OF SELF-ESTEEM

 

 

 

 

 

 

 

 

 

(N=42)
1. ANOVA Results
Source d.f. SS MS F sig
Assessment (A) 1 3.5323 3.5323 6.5221 0.013
Self-Esteem (SE) 1 0.9421 0.9421 1.7027 0.196
A x SE 1 2.8515 2.8515 5.2650 0.024
Before-After (BA) 1 0.0426 0.0426 0.0787 0.780
A x BA 1 0.1565 0.1565 0.2890 0.592
SE x BA 1 0.7907 0.7907 1.4600 0.231
Error 77 41.7022 0.5416
Total 83 52.5870
11. Mean Scores for each Group in ANOVA
Before After
Assessment Assessment
Lo(:) Hi(+)_ Lo(:) Hi(+)
HSE 7.5 7.0 7.2 7.3
Self— (N= 6) (N=ll N= 6) (N=ll)
Esteem LSE 6.5 7.5 6.5 7.1
N=15) (N=lQ) N=15) (Nslo

 

 

 

 

III. Scheffe Comparisons

 

(p j .10)

 

 

 

 

1. Assessees who received high assessment ratings received signifi-
cantly higher supervisor job performance ratings than did
assessees who received low assessment ratings.

2. High self-esteem assessees with high assessment ratings received
significantly higher supervisor job performance ratings than did
low self-esteem assessees with low assessment ratings.

3. No other differences between all possible combinations of
means were significant.

77

The assessment variable effect on the supervisor job
performance rating was significant difference at the .01
level. The Scheffe tests confirmed that there was a sig-
nificant between mean job performance rating Of assessees
who received above median assessment ratings and the mean
performance rating of those assessees who received below
median assessment ratings. This finding suggests that
there was a significant agreement between assessors and
supervisors as to the "high" and "low" performers (Table 3—1).

Also a significant F ratio (.02) was present for the
interaction of assessment and self-esteem (A x SE). The
Scheffe test confirmed that there was a significant dif-
ference in mean job performance ratings between the HSE+

group of assessees and the LSE- group (Table 3-1).

Impact of Assessment on Satisfaction with Work

Table 3-2 reports the ANOVA results with respect to
job satisfaction (work scale), examining the relationship
between the dependent variable satisfaction with work and
the main effects of assessment ratings, self-esteem, and
before-after. For this analysis, data were available for

32 subjects who completed the JDI during assessment and

again six months after assessment.

RELATIONSHIP BETWEEN:

78

TABLE 3-2

(A) BEFORE AND AFTER ASSESSMENT, (B) JOB

SATISFACTION (WORK), AND (C) ASSESSMENT RATING, FOR ASSESSEES

AT DIFFERENT LEVELS OF SELF—ESTEEM

 

 

 

 

 

 

 

 

 

(N=32)
1- ANOVA Results
Source d.f. 88 MS F sig
Assessment (A) 1 96.4405 96.4405 2.4513 0.123
Self-Esteem (SE) 1 258.2827 258.2827 6.5649 0.013
A x SE 1 11.5030 11.5030 0.2924 0.591
Before—After (BA) 1 19.8376 19.8376 0.5042 0.481
A x BA 1 11.5836 11.5836 0.2944 0.590
SE x BA 1 15.2542 15.2542 0.3877 0.536
Error 57 2242.5393 39.3428
Total 63 2609.7500
11. Mean Scores for each Group in ANOVA
Before After
Assessment Assessment
Lo(e) Hi(f), ,Lo(r) Hiji)
Self— HSE 42.1 37.5 2.6 40.4
Esteem N=7), (N=11 N=7) ,(N=ll)
LSE 35.8 37.8 38.8 33.4
(N= 9) (*2 N=9) OPS)

 

 

 

 

 

 

 

 

III.

Scheffe Compprisons (p f .10)

1. High self-esteem groups were significantly more satisfied with work
than were low self-esteem groups.

2. No other differences between all possible combinations of
means were significant.

79
The self-esteem effect was significant at the .01
level, indicating a tendency for high self-esteem persons
to be more satisfied with their work than low self-esteem
persons. The Scheffe test confirmed that the two groups
of HSE assessees were significantly higher than the two
LSE groups of assessees in their satisfaction with work

(Table 3-2).

Impact of Assessment on Satisfaction with Supervision

Table 3-3 reports the ANOVA results with respect to
job satisfaction (supervision scale), examining the rela-
tionship between the dependent variable satisfaction with
supervision and the main effects of assessment rating,
self—esteem, and before-after. For this analysis, data
were available for 32 subjects who completed the JDI

during assessment and again six months after assessment.

RELATIONSHIP BETWEEN:

80

TABLE 3-3

(A) BEFORE AND AFTER ASSESSMENT, (B) JOB
SATISFACTION (SUPERVISION), AND (C) ASSESSMENT RATING, FOR
ASSESSEES AT DIFFERENT LEVELS OF SELF-ESTEEM

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(N=32)
1. ANOVA Results
Source d.f. SS MS F 813
Assessment (A) 1 98.2572 98.2572 1.7974 0.185
Self-Esteem (SE) 1 5.7784 5.7784 0.1057 0.746
A x SE 1 339.5707 339.5707 6.2118 0.016
Before-After (BA) 1 5.4383 5.4383 0.0995 0.754
A x BA 1 0.1665 0.1665 0.0030 0.956
SE x BA 1 0.0320 '0.0320 0.0006 0.981
Error 57 3115.9277
Total 63 3621.7344
II. Mean Scores for each Group in ANOVA
Before After
Assessment Assessment
Lo(:) Hi(:) Lo(:) Hi(})
HSE 50.9 42.7 50.6 43.9
Self— _(N=7) (N=ll N=7) (N=ll)
Esteem
LSE 44.6 48.0 46.0 47.0
N=9) (N=5) 4(N=9) (N=5)
III. Scheffe Comparisons (p f .10)

 

1. Low self-esteem assessees who received high assessment ratings
and high self-esteem assessees who received low assessment ratings
were significantly more satisfied with supervision than were high
self-esteem assessees who received high assessment ratings and
low self—esteem assessees who received low assessment ratings.

2. No other differences between all possible combinations of

means were significant.

81
The interaction of assessment and self-esteem
(A x SE) was significant at the .02 level. The Scheffe
test confirmed that the LSE+ and HSE- groups of assessees
were significantly more satisfied with their supervision

than the HSE+ and LSE- groups (Table 3-3).

Impact of Assessment on Satisfaction with Promotions

 

Table 3-4 reports the ANOVA results concerning job
satisfaction (promotions scale), examining the relation-
ship between the dependent variable satisfaction with
promotions and the main effects of assessment rating,
self-esteem, and before-after. For this analysis, data
were available for 32 subjects who completed the JDI

during assessment and again six months after assessment.

 

RELATIONSHIP BETWEEN:

82

TABLE

3-4

(A) BEFORE AND AFTER ASSESSMENT, (B) JOB
SATISFACTION (PROMOTIONS), AND (C) ASSESSMENT RATING, FOR
ASSESSEES AT DIFFERENT LEVELS OF SELF-ESTEEM

 

 

 

 

 

 

 

 

 

(N=32)
I. ANOVA Results
Source d.f. SS MS F sig
Assessment (A) 1 13.5951 13.5951 0.2510 0.618
Self-Esteem (SE) 1 39.1862 39.1862 0.7236 0.399
A x SE 1 72.0675 72.0675 1.3308 0.253
Before—After (BA) 1 274.4924 274.4924 5.0687 0.028
A x BA 1 36.1126 36.1126 0.6669 0.418
SE x BA 1 26.2778 26.2778 0.4852 0.489
Error 57 3086.7784 54.1540
Total 63 3576.8594
x1, Mean Scores for each Group in ANOVA
Before After
Assessment Assessment
Lo(j) Hi(+ Lo(-) Hi(+)
HSE 21.6 20.2 13.3 17.1
Self— N=7 N=11) N31) (N=ll
Esteem
LSE 23.0 19.6 19.6 16.6
(N=9) (N151 (N= 9) (N=5)

III.

 

 

 

 

Scheffe Comparisons

(p:

 

 

 

 

.10)

1. High self-esteem assessees who received low assessment ratings
were significantly lower in satisfaction with promotions after

assessment .

2. No other differences between all possible combinations of

means were significant.

83
The before-after (time) effect was significant at
the .03 level. The Scheffe test confirmed that there
was a significant decrease after assessment in satisfac-
tion with promotions on the part of the high self-esteem

assessees who received low assessment ratings (Table 3-4).

Impact of Assessment on Satisfaction with Co-Workers

 

Table 3-5 reports the ANOVA results with respect to
job satisfaction (co-workers scale), examining the rela-
tionship between the dependent variable satisfaction with
co—workers and the main effects of assessment rating,
self-esteem, and before-after. For this analysis, data
were available for 32 subjects who completed the JDI

during assessment and again six months after assessment.

 

RELATIONSHIP BETWEEN:

84

TABLE

3-5

(A) BEFORE AND AFTER ASSESSMENT, (B) JOB
SATISFACTION (CO-WORKERS), AND (C) ASSESSMENT RATING, FOR
ASSESSEES AT DIFFERENT LEVELS OF SELF-ESTEEM

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(N=32)
1. ANOVA Results
Source d.f. SS MS F sig
Assessment (A) 1 67.9095 67.9095 1.2599 0.266
Self-Esteem (SE) 1 66.1171 66.1171 1.2266 0.273
A x'SE 1 61.7411 61.7411 1.1454 0.289
Before—After (BA) 1 17.3040 17.3040 0.3210 0.573
A x BA 1 9.2326 9.2326 0.1713 0.681
SE x BA 1 5.4915 5.4915 0.1019 0.751
Error 57 3072.3882 53.9015
Total 63 3370.0000
II. Mean Scores for each Group in ANOVA
Before After
Assessment Assessment
Lo(j) H (+) Lo(r) Hi(+)
HSE 49.1 44.5 46.7 43.0
Self— (N=7) .1N=ll) (N=7) (Bell)
ESteem LSE 47.3 47.6 48.8 46.2
1N= 91 (N=5L (N=2 (Na-5)
Ill. Scheffe Comparisons (p f .10)

 

1. No differences between all possible combinations of
means were significant.

85
There were no significant effects found in this par-

ticular analysis of variance (Table 3-5).

Relationships of Ghiselli SDI Traits
In this section results are reported of the cor-
relational analyses of the relationships of the 13 Ghiselli
SDI traits with: (1) job performance; (2) job satisfac-

tion; and (3) assessment ratings.

Ghiselli Traits and JOb Performance

The correlations of the 13 Ghiselli SDI traits with
the three measures of before assessment jOb performance
are reported in Table 3-6. Fifty—six (56) subjects re-
sponded to the Ghiselli SDI and completed a self—rating
form, as part of the assessment battery of tests. These
56 subjects also were rated by two co—workers and their
supervisor before assessment results were known. For
Iourposes of this analysis, co—worker ratings were averaged

:Eor a "peer" rating.

CORRELATIONS OF GHISELLI TRAITS WITH OVERALL JOB
PERFORMANCE RATINGS OBTAINED BEFORE ASSESSMENT
(N=56)

86

TABLE 3-6

 

Ghiselli Trait

Performance Ratings Correlations

 

Peer Supervisor Self
Supervisory Ability ~.34* -.1O .31*
Intelligence -.06 .07 .46**
Initiative .02 .02 .28*
Self-Assurance —.12 .04 .38**
Decisiveness .08 .14 .16
Masculinity-Femininity -.13 .09 .19
Maturity —.12 -.10 .04
Working Class Affinity .05 -.03 .06
Achievement Motivation .01 .06 .37**
Need for Self-Actualization .00 -.ll .22
Need for Power .10 .08 .39**
Need for High Reward -.01 .05 —.13
Need for Security -.06 .02 -.46**

 

* Significant at .05 level
** Significant at .01 level

 

 

87
The Ghiselli SDI characteristics had no significant

correlation with supervisor ratings. with the exception

 

of the supervisory ability trait, the SDI characteristics

had no significant correlation with pger ratings (Table

 

3-6).

Seven of the SDI traits did have significant correla- *
tions with self—ratings. This suggests that the assessee
has a fairly constant image of himself as measured by two

assessment instruments, one evaluating his self-concept of

 

job performance and the other his concept of his person-

ality (Table 3-6).

Ghiselli Traits and Job Satisfaction

The correlations of the 13 Ghiselli SDI traits with
the four measures of before assessment job satisfaction
are reported in Table 3-7. These correlations were ob-
tained from the JDI administered to 34 subjects as part
of the assessment battery of tests. Four scales of the
JDI were used to measure the assessees' attitude toward

the job.

CORRELATIONS OF GHISELLI TRAITS WITH JOB SATIS-

88

TABLE 3-7

FACTION SCALES OBTAINED BEFORE ASSESSMENT

(N=34)

 

 

Job Satisfaction Scales Correlation

 

 

Ghiselli Traits Co-
Work Supervision Promotions Workers

Supervisory Ability .24 -.17 .12 .01
Intelligence .15 .09 -.14 .05
Initiative .30 -.13 .03 .12
Self—Assurance .22 .03 -.14 .07
Decisiveness .31 .09 .19 .12
Masculinity—Femininity .09 -.26 -.16 .23
Maturity .30 -.04 -.04 .24
Working Class Affinity .20 -.10 .18 .07
Achievement Motivation .30 .03 .05 .22
Need for Self-Actual-

ization .02 -.23 .12 -.09
Need for Power .16 .37* -.24 -.07
Need for High Reward .37* .01 -.O7 -.18
Need for Security .27 -.00 -.12 -.11

 

* Significant at .05 level

89
Only two traits had significant correlations with
job satisfaction: (1) need for power with the supervision
scale; and (2) lack of need for high financial reward with
the work scale (Table 3—7). In essence then, the SDI had
virtually no relationship with the job satisfaction mea-

sures .

Ghiselli Traits and Assessment Ratings

 

Correlations of the 13 Ghiselli traits with assess-
ment ratings are reported in Table 3-8. For this analysis,

assessment ratings for 56 subjects were used.

9O

coHumofioaano cmuuwuz
oouumoficsaaoo Hmuo
unmaanmowaa

mwcmno Cu owcoamwm
m>HumHuficH

.oa

\ONQO‘

coaummoaon

huwawn< HmGOHumecmmuo

mmoco>fimaooa

wafixmz ooamaomn
mfinmuwwmmq

umeflmHHW> UCwEwwmmg +

qulfi

.H

HmsmH Ho. 06 0:60HHchHm «4
H6>6H mo. 0m 0amUHHHame 4

 

 

m~.- 4H.- H~.- .«mm.- «.Hm.- 4Hm.- s~.- 44Hs.- mH.- «.me.- «46m.- suHusumm 06H ewmz
so.- no. No.1 mH.- oo.- no.1 HH.- SH.- ~o.- NH.- ao.- eumzmm :Hm anm you 6662
mm. mo. mo. wH. NN. mo. NH. «Hm. mH. «N. aH. 06368 06H emmz
HH. mo.u «a. Hm. ow. mH. MH. SN. OH. om. mH. =6Hum~HHmnuu<IHH6m How emmz
««mm. «mm. «om. «8mm. aww. «N. *Hm. acm. «mm. ««mq. «yum. coaum>fiuoz uaoam>wano<
mo.- ~H.- SH.- 6H.- mH.- H~.- HN.- oH.- NN.- mH.- mH.- HuHcHHH< mmmHo maquoz
mo.- No.1 ao.- no.1 mH.- om.u mH.- no.1 mo.- HO.. OH.. suHusumz
SH. ao. mH. mo. mo. HH. HH. oH. oH. «N. «H. suHaHaHa6a1s0HaHHaummz
Ho. HH. mo. mH. om. mo. OH. mH. mo. SH. SH. mmmcm>HmHoma
«on. 50. mm. 4mm. aH. NN. cm. «H. mm. «44m. «mm. moamusmm<uuHmm
AH. HO. so. SH. SH. aH. 8H. mm. oH. 4cm. mH. 6>H08H0H6H
«ans. mm. ««m¢. ««Hm. ««mm. «amm. «8mg. «*08. *«qq. ««mq. «ems. mocmwfiaaoucH
mo. mo.- mH. NH. mH. HH. mH. 6H. so. 6H. NH. a0334 spomH>Hmaam

+oH a m a o m e m N H HHm
num>o onue HHHmmHeu

 

cowumamuuou mdwumm ucmammmwm<

 

 

Homuzv

mquH<m Hszmmmmm< :BHB mHH<MH HAQMmHmU mo mZOHH<AmmmOU

mum mqm<e

91

Four traits had significant correlations with overall

easssessment rating (Table 3-8). Of the four, intelligence

'kuad highly significant correlations with nine of the ten

iJndividual assessment ratings (Table 3-8). This suggests

'that intelligence, as measured by the SDI, was the trait
Inost highly correlated with assessment ratings. That is,
‘the more intelligent the assessee, the better he did at
the assessment center.

For the other three traits with significant correla-
tions with the overall assessment rating: achievement had
significant correlations with nine individual ratings;
self-assurance (self-esteem) had three; and lack of need
for security had significant correlations with five indi-
\/idua1 assessment ratings (Table 3-8).

There was a significant correlation between the trait

c>f initiative and the assessment rating on the leadership

\rariable (Table 3-8).

§EEEE£X
The basic results and findings of the statistical

analyses of the research data were presented in this

chapter. The next chapter presents a more extensive dis—

cussion of the research findings.

CHAPTER 4

SUMMARY, DISCUSSION, AND CONCLUSIONS

This research has been concerned primarily with the

 

effects of a personnel assessment experience, which could
be a truly significant event in an individual's organiza-

tional life, with respect to its impact on his job perform-

 

fi
ance and job satisfaction.

It is logical to assume that assessee reaction to
assessment will differ and that these differences might be
accounted for at least in part by personality character-
istics, particularly self-esteem. This logical assumption
led to the develOpment of the hypotheses presented in
Chapter 2. Basically these state that after an assessment

center experience, job performance and job satisfaction

would:
1. Increase for those assessees with high self—
esteem who attained above median assessment
ratings —- the HSE+‘group of assessees.

2. Not change for those assessees with high self-
esteem who attained below median assessment
ratings -- the HSE- group of assessees.

3. Not change for those assessees with low self-
esteem who attained above median assessment
ratings -- the LSE+ group of assessees.

92

93
4. Decrease for those assessees with low self-
esteem who attained below median assessment
ratings -- the LSE- group of assessees.

The subjects were 60 managers who participated in a
personnel assessment center, conducted by the parts divi-
sion of a manufacturing firm in the automotive industry.
As part of the assessment process, assessment ratings
were obtained on ten variables for each subject. These
were combined for use as a dependent variable here. Addi-
tional data, identified by the company research staff as
being gathered for research purposes, were also collected
for the present study. These data were measures of:

1. ‘Personality characteristics, where the

measuring instrument used was the Ghiselli
Self-Description Inventory (SDI). The SDI
was administered as part of the assessment
center battery.

2. Job performance, where a supervisor's rating

was used. A rating was obtained on the
assessee before assessment and again about
six months after assessment.

3. Job satisfaction, where the measuring instru-

ment used was the Job Descriptive Index (JDI).
The JDI was administered as part of the
assessment center battery and again about

six months after assessment.

To test the hypotheses, a three—way analysis of vari-
ance was used. That is, a 2 x 2 x 2 factorial design was

employed. In the design, the main effects were: (1)

assessment rating_(above median or +, below median or -):

 

94
(2) level of self-esteem (above median or HSE, below median
or LSE); (3) Elma (before assessment, after assessment).
The dependent variables were job performance (supervisor's
rating) and different aspects of job satisfaction (JDI
scores). Where the analysis indicated a significant F
ratio, the Scheffe method of means comparison was used to
test the significance between all possible combinations of
pairs of means.

Adjunct to the primary purpose of the research was a
determination of the relationship between personality
characteristics and two types of criteria: performance
ratings and satisfaction scores. The Ghiselli SDI mea-
sured personality characteristics. Performance ratings
used were self, peer, and supervisor ratings completed
before assessment results were known. Satisfaction scores
were from work, supervision, promotions, and co-workers
scales of the JDI administered before assessment results
were known.

The relationship between personality characteristics
and assessment center evaluation was also examined. The
Ghiselli SDI was used to determine personality character-
istics. The assessment center staff's ratings of ten
variables were used as a measure of assessment. Correla-

tions between SDI traits and assessment ratings were

95

obtained.
The results of the statistical analyses are reported

in the preceding chapter. These will be interpreted and

discussed in this chapter.

The Impact of Assessment on
Performance and Satisfaction

The analysis of variance led to the rejection of two
hypotheses, partial support for one hypothesis, and sup-

port for one hypothesis.

1. No significant changes were found in after
assessment job performance and job satisfac-
tion for the HSE+ assessees. Thus this
hypothesis was rejected.

2. No significant changes were found in after
assessment job performance for the HSE-
assessees. This finding supports the
hypothesis. But a significant decrease was
found in satisfaction with promotions after
assessment. The change in job satisfaction
is contrary to the effect hypothesized.
Accordingly, the hypothesis is supported in
part and rejected in part.

3. No significant changes were found in after
assessment job performance and job satisfac-
tion for the LSE+ assessees. These findings
support the hypothesis.

4. No significant changes were found in after
assessment job performance and job satisfac-
tion for the LSE— assessees. Thus this
hypothesis was rejected.

In general then, results of the analysis lead to a

conclusion that for the most part an assessment center

96
experience does not affect assessee job performance, at
least in the time span measured. It also appears that

assessment does not affect job satisfaction, except perhaps

 

in the case of satisfaction withppromotions.

For satisfaction with promotions, the findings indi-
cated that the high self—esteem individuals who received
low assessment ratings (HSE- assessees) became signifi-
cantly less satisfied with promotions after assessment
(Table 3-4). The finding may indicate an underlying dis—
satisfaction with promotions which has come to the surface
after the assessment center experience. If so, there may
be many reasons for the dissatisfaction. To uncover the
reasons would require further in-depth research. From the
data, it seems reasonable to conclude that the dissatis-
faction with promotions may arise because the HSE-
assessees are disappointed by their low assessment ratings.
Such ratings would conflict with their self-perceived
competence, and might be regarded as a "failure". Since
an HSE person tends to externalize failure, he may blame
low ratings on the assessment process. In turn, since they
may relate the assessment program with promotions, this
could lead to more dissatisfaction with promotions.

This finding could be of significance to managers.

It indicates that perhaps concern should be directed to

97
the HSE person who does poorly at an assessment center.
This could have important implications, for as shown in
Table 3-1, supervisors tend to rate the job performance
of high self—esteem assessees with low assessment ratings
(HSE- group) as being basically the same as the job per-
formance of those who do well in assessment (HSE+ and LSE+

groups). Thus it appears that sppervisors regard the HSE-

 

individual who gets a low rating in assessment as a "good

 

performer". Therefore, it is reasonable to expect that

it may be desirable to retain this individual, and not
lose him because of dissatisfaction stemming from his poor
assessment showing.

One way to maintain the HSE- assessee's job satisfac-
tion is to "tailor" the feedback of assessment results.
Such feedback should stress the importance of a develop-
ment program to overcome weak areas. It should minimize
the impact of assessment results on future promotions.

For example, the assessee could be told that while the
assessment results are undoubtedly disappointing, it is
better to uncover weaknesses at this point in a career,
rather than later. Now they can be corrected if effort
is devoted to a development program. If he were promoted
and then the managerial weaknesses showed up, he may

suffer embarrassment and frustration, and perhaps even a

98
demotion. If possible, the experiences of other HSE-
assessees could be used. For example, other HSE- assessees
may have experienced similar assessment "shock": but these
persons may have undertaken develOpment programs and sub-

sequently continued to progress in the organization.

Lack of Significant Findings

The lack of statistically significant findings sup-
porting the hypotheses may be due to several factors.

1. The measuring instruments may not have been
sensitive, so the lack of significant findings
stem from measurement error.

However, the SDI (used to measure personality charac-
teristics) and the JDI (used to measure job satisfaction)
were reliable and significantly related to other measures
(Tables 2-3 and 2-4). The supervisor job performance
rating (used to measure job performance) had a high
internal reliability and a moderate degree of correlation
with two other measures of job performance: the peer—
average rating and the self—rating (Tables 2-7 and 2-9).
The composite assessment rating used also had a high
degree of internal reliability. It appears then that the
measuring instruments were adequately reliable for research

purposes.

99

2. The measuring instruments may not have

been taken seriously by the assessees
and their supervisors.

The instruments were identified as being part of a
company research program to evaluate the assessment pro-
gram. For this reason, assessees and supervisors might
not give serious consideration to completion of the
instruments. Or they might tend to "slant" their answers,
based on their attitudes toward the assessment program.
This is a problem confronted in many research projects.
It is a problem which is extremely difficult to overcome.
Basically, the researcher must rely on his subjects
making an honest effort in completion of measuring

instruments.

3. The small number of subjects in the research
sample may have precluded significant findings.

In this research, the small number of subjects may
well have limited statistical significance. With a small
number of subjects involved, large changes in job per-
formance and job satisfaction would have to be observed
in order for the changes to be statistically significant.
Further, with a small sample, individual differences may
have a major effect on cell means. For example, there
were only 5 assessees in the LSE+ group for job satisfac-

tion analyses. If one assessee reacted to assessment

100

differently from the other four in the group, the mean
for the group would change considerably. With a large
number of subjects, individual differences would not
impact on the group mean. Also, with a large number of
subjects, small changes in performance and satisfaction
would likely be more statistically significant wherever ET
they occur. Unfortunately, sample size is often a problem

with this kind of research. Rather than forego research,

 

it seems advisable to proceed even with a small group of
subjects.

4. The time span between measures may not have
been such that any effects were observable.

The time dimension could be extremely important.
For example, there well may be an immediate effect on job
performance and job satisfaction after feedback of assess-
ment results. This effect may not be present six months
later, when the after assessment measures in this research
were taken. Or, there may be a long-range effect on per-
formance and satisfaction if the assessee becomes con-
vinced that the assessment results are indeed affecting
his career with the organization.

5. The moderators selected may not be important
as they relate to the effects of assessment.

This research suggests that self—esteem does not

moderate the effects of assessment on subsequent job

101
performance and job satisfaction. But other variables
might. In Table 3-6 it is noted that three personality
traits, other than self-assurance (self-esteem) have
significant correlations with assessment ratings.
Certainly the search for moderating variables should not
be abandoned.

6. Assessment has limited or no impact on an
individual.

The reason for this limited or no impact is that an
assessment center experience does not affect assessee job
performance and job satisfaction. A possible explanation
for this conclusion may lie in the nature of the assess-
ment center used in the research. By stressing assessment
as a tool for develOpment, and deemphasizing it as a tool
for promotion, the organization may have convinced
assessees that assessment was not a "threat" to their
careers. Consequently, assessees had no reactions to
assessment, except for the HSE— assessees who were appar-
ently disturbed by the fact that they did not do as well

as they had anticipated.

_§ignificant Findings

 

Now turning to a discussion of what the research did
Jreveal, several statistically significant findings did

emerge from the analyses of data.

102

1. Assessees who attained high assessment

ratings received higher before-after
supervisor job performance ratings than
those assessees who attained low assess-
ment ratings (Table 3-1).

This finding indicates that there is agreement
between assessors and supervisors as to which group of
assessees were ”high performers" and which were "low
performers". The finding suggests that assessors and
supervisors, using the same set of ten items on a rating
scale, are able to identify relative effectiveness in a
similar fashion, independent of the situation in which
they are rating. Supervisors saw assessees perform over
a period of time, and rated them on differing sets of
tasks than assessment staff would. In assessment, indi-
viduals performed for a short time and were rated on a
standard set of tasks. Despite these different rating
situations, both supervisors and assessors agree as to
who are "high performers" and as to who are "low per-
formers".

The close agreement implies that perhaps an assess-
ment center may not be a necessary method for identifying
management potential. If so benefits other than identifi-
cation of management potential should be derived from

assessment for the assessment program to be worthwhile.

One such benefit may be that assessment reports provide

 

if!) -

103
data from which recommendations for assessee develOpment
can be made. Another benefit may be that assessment pro-
vides a standardized evaluation of individuals from all
parts of the organization. That is, without assessment
it would be difficult to evaluate management potential by

comparing supervisor ratings of those who work in sales

offices with ratings of those who work in parts depots.
Perhaps the most convincing argument for the assess—
ment center is that, despite the findings here, only a .f
I. 5'1
L.»

 

moderate correlation (r = .30, p i .05) exists between
assessment ratings and before assessment supervisor per-
formance ratings (Table 2—9). This correlation suggests
that the assessment center may be evaluating different

aspects of management potential, which supervisors are

unable to discriminate.

2. High self—esteem persons who attain high
assessment ratings (HSE+ assessees)
received higher before-after supervisor
job performance ratings than did low self-
esteem persons who attained low assessment
ratings (LSE- assessees) (Table 3-1).
This finding is another indication of the agreement
between assessors and supervisors as to who are "high
performers" and who are "low performers". But this agree—

ment is not complete, which may account for the moderate

correlation discussed under the preceding finding. For

104

’example, as shown in Table 3-1, supervisors and assessors
disagreed as to the "performance" of the HSE- group of
assessees. Supervisors considered the group to be "high
performers". Assessors considered the group to be "low
performers”.

The implication of this finding is that there may be F5
a group of assessees, who do poorly at assessment, whose
loss to the organization may not be detrimental. This

group are the LSE- assessees. Both assessors and super-

 

visors agree that these assessees are "poor performers". '
This is the only group of assessees for which such agree—
ment exists.

The pattern of agreement between assessors and super-
visors suggests that perhaps the concern for the person
who does poorly at an assessment center should be directed
to the HSE person, and not to the LSE person as conceptual-
ized in Chapter 2.

3. Assessees with high self—esteem were more

satisfied with their work than assessees
with low self-esteem (Table 3-2).

The explanation for this finding may lie in the very
concept of self-esteem. That is, the HSE individual per-
ceives himself as being competent to perform virtually any

task. In essence, he enjoys his work and consequently he

is satisfied with his work environment. On the other hand,

105

the LSE individual is less confident of his ability to

perform tasks. He may even dread his work and consequently

be less satisfied with the work environment than his HSE

fellow-worker. From the finding, it appears that the HSE

person likes his work, while the LSE person does not.

4. Low self-esteem assessees who attained high
assessment ratings (LSE+ group) and high
self-esteem assessees who attained low
assessment ratings (HSE- group) were more
satisfied with supervision than high self-
esteem assessees with high assessment
ratings (HSE+ group) and low self-esteem
assessees with low assessment ratings (LSE—
group) (Table 3—3).

There are little data available which help to explain

this finding.

However, some implications may be drawn.

Since the finding is complex, it is necessary to consider

its implications for each assessee group separately.

a.

HSE+ assessees appear dissatisfied with
the style of their supervisors, although
they are considered "high performers" by
both supervisors and assessors. The dis-
satisfaction may be a result of the
subjects' high self-esteem level which
leads to a feeling that supervisors, in
general, are less competent than they.

HSE- assessees appear to like their super-
vision, as they liked their work. Their
satisfaction in these two areas contrasts
with their lack of satisfaction with pro-
motions, which was discussed earlier in
the chapter. It may be that the "threat"
to their career, they may have perceived
arising from their low assessment ratings,
has had an impact on their thinking only

with respect to promotions and not to
other aspects of the job.

106

c. LSE+ assessees appear to like their
supervisors' style. Perhaps the super—
visor would want to be aware that there
are some ”good" performers who, because
of their low self—esteem, may need
encouragement.

d. LSE- assessees appear to be a group that
is not going anywhere in the organization.
All indications are that they are "poor"
performers. These assessees apparently T,
are not satisfied with their supervision “a
and their work. Supervisors may want to
take a "hard look" at this group of indi-
viduals.

 

Relationships of Ghiselli SDI Traits k5

 

The relationships of the SDI personality character-
istics with job performance ratings, job satisfaction
scores, and assessment ratings can be discussed briefly

because only a few significant relationships were found.

Performance Ratings

 

The findings indicate that virtually no relationship

exists between the SDI and peer-averagegperformance

 

ratings and the SDI and sppervisor performance ratings

 

(Table 3—6). However, there are a number of significant

correlations between SDI traits and self-ratings of per-

 

formance (Table 3-6). Such correlations could occur
because both are self-ratings. It is to be expected that
an individual would have a constant image of himself which

might bias his response to all the instruments.

 

107

The latter correlations reveal a pattern of person-
ality characteristics similar to those found in Table 3—8.
They indicate that the self-rated "high performer" has
traits of intelligence, achievement motivation, and lack
of need for security. In addition, the individual who
rated himself a "high performer" revealed strong traits of
self-assurance and lack of need for power. To a lesser
extent he revealed traits of supervisory ability and
initiative.

The correlations among SDI traits and self-ratings
(Table 3—6) suggests that the individual has a fairly
constant image of himself as measured by two instruments,
one determining a variety of personality traits, the other

a self-concept of job performance.

Satisfaction Scores

 

The findings indicate that virtually no relationship
exists between the SDI and the work, supervision, pro-
motions, and co—workers scales of the JDI, despite the
fact that both are self—ratings. In essence, the findings
are inconclusive, although they perhaps may suggest that
one's self-estimate of personality characteristics has no
relationship with one's attitude toward various facets of

his job.

108

Such results are possibly not consistent with Eran
(1966). He used the decision making approach (now
decisiveness) scale of the SDI to separate managers into
"highs" and "lows". Using the Porter Management Positions
Questionnaire to measure job attitudes, Eran found that
those with high self—ratings of decisiveness were sig-
nificantly more satisfied with their job as managers.

In the present research it was found that high self-
esteem assessees were significantly more satisfied with
their work than were low self-esteem assessees. But this
is only one significant correlation out of a possible six

on this satisfaction scale.

Assessment Rapings

The personality characteristics of the individual who
does well at an assessment center do emerge from the data.
For as shown by the correlations, the assessee who attains
high assessment ratings describes himself as intelligent,
achievement motivated, and lacking in need for security.
To a lesser extent he is also self-assured.

Of the four traits just cited, self-ratings of
intelligence appear to be the most highly correlated with
assessment results. In turn, assessment results are pre-

dictive of management potential, as discussed in the review

109
of the research literature in Chapter 1. Thus the intel-
ligent scale may be most useful in predicting management
potential. This view is supported by such researchers as
Stogdill (1948), Goode (1951), Randle (1956), and Harrell
(1961) who indicate that intellectual ability is an im-
portant trait in successful managers.

However, the findings do also indicate some disagree-
ment between the assessment ratings and the Ghiselli SDI
over what appear to be similar variables. There is no
correlation between some variables where, at least on the
face of it, there should be a correlation. For example,
both the assessment staff and the SDI purport to measure
"initiative" and "decisiveness". But the findings (Table
3-8) indicate that these traits as viewed in assessment
are not the same as measured by the SDI. "Leadership"
and "supervisory ability" connote similar abilities, but
the findings (Table 3-8) indicate that they are not the
same. Likewise, "organizational ability” and "supervisory
ability" connote similar abilities, but again the findings
(Table 3-8) indicate that they are not the same. This
suggests the need for better conceptualization of the
meaning of the characteristics rated.

While the assessment center and the SDI both attempt

to identify management potential, both may be looking at

 

 

110
different slices of this potential. Hence, one method can-
not be readily substituted for the other, and still obtain
the same assessment results. In addition, as suggested
Ipreviously, assessment provides identification of manage—
Inent potential for a diversified number of candidates on a
similar set of tasks. Assessment also provides information
tiseful to the candidate for develOpment and to the organ-

ization in providing develOpmental opportunities.

Implications for Management
The results of the present study indicate that manage—
Inent should be aware of the impact an assessment program
nuay have upon individuals and the organization. The
Etnalysis suggests that there are some implications to be
noted.

1. High self-esteem persons who received low
assessment ratings are rated high by super-
visors (Table 3-1). This group became
significantly less satisfied with promotions
after assessment (Table 3-4). But they were
significantly satisfied with work and super-
vision (Tables 3-2 and 3-3).

Should this dissatisfaction remain high, it could

Ileuad to turnover of those regarded as "good" performers.
'Tfue pattern of satisfaction may indicate that this group

"fiits" the organization despite their poor assessment

Srudwing. Hence, a development program may prove beneficial

111
to both the individual and the organization.

2. Low self-esteem persons who received low
assessment ratings are rated low by super-
visors (Table 3-1). This group indicated
dissatisfaction with work and with super-
vision (Tables 3-2 and 3-3).

This group of assessees appear low on nearly all mea-
sures. Consistent low rankings, in comparison with the
<3ther three groups of assessees, suggests that this par-
ticular group should be evaluated. In other words, a

"good hard look" may identify some individuals who might
loe more effective with a job or career change.

3. Low self-esteem persons who received high
assessment ratings are rated high by super-
visors (Table 3-1). This group indicated
dissatisfaction with work (Table 3-2), and
satisfaction with supervision (Table 3-3).

This reflection of dissatisfaction and satisfaction
nuay indicate that one group of "good" performers needs
fiirther praise and encouragement for their work efforts
fdrom.supervisors. Such action may enhance the groupHs

OVerall job satisfaction.

4. High self-esteem persons who received high
assessment ratings are rated high by super-
visors (Table 3-1). This group indicated
dissatisfaction with supervision (Table 3-3),
and satisfaction with work (Table 3-2).

This reflection of dissatisfaction and satisfaction

Tuadz indicate that another group of "good" performers

ineueds a modification in the management style to which

112

they are exposed. This may occur through training for

their supervisors, or perhaps with a job change.

5. There was significant correlation (r = .30,
p S .05) between assessors and supervisors
over "high performers" and "low performers"

(Tables 2—9 and 3-1).

The agreement between supervisors and assessors may

raise questions concerning the need for an assessment

program. However, it should be recognized that benefits

over and above identification of "good" and "poor" per-

formers may accrue from an assessment program. For ex-

ample, assessment provides information which in some

organizations has proved useful in establishing develOp—

ment programs for assessees.

Suggestions for Future Research

It is suggested that this research be replicated. A

larger sample size would be desirable. Also, consideration

Inight be given to using an assessment center where the

eemphasis is on the use of assessment results for future

forogression in the organization. Where such an objective

eocists, findings may be different from those in the present

stuady where the assessment center's objective was supposed
tc> be develOpment of the assessee.
Different moderating variables may also result in

diffiferent findings. For this reason, it is suggested that

113

traits determined by instruments other than the SDI be

used as possible moderators of the effects of assessment

ratings on assessee job performance and job

satisfaction.

These moderators could be examined singly and in combina-

tion to determine if any effects different from those in

the present research are present.

The time dimension should be more controlled. In the

present research measures were obtained at two points in

time. A more comprehensive longitudinal study, with mea-

bur—"H.

sures at several points in time, may reveal effects of

assessment ratings on assessee job performance and job

satisfaction.

For any future research, the same research design is

believed adequate. That is, a 2 x 2 x 2 factorial design

appears suitable for analyzing the research
a repeated measure design (the before—after
the repeated variable) should be considered

provide more complete analysis of the data,

data. However,
time would be
since it may

in the sense

that both within and between subject variance would be

accounted for.

Conclusion

 

The reported research has attempted to

determine the

extent to which participation at an assessment center

1

114

affects assessee job performance and job satisfaction.

The research findings indicated that with one exception,

no significant changes occurred in assessee job perform-
ance and job satisfaction after assessment. The one ex-
ception noted was that high self-esteem assessees who
received low assessment ratings (HSE- group) were sig-
nificantly less satisfied with promotions after assessment.

Accordingly, the management concern mentioned in the
second chapter, for the assessee who does poorly at an
assessment center, appears to be an apprOpriate concern
for the HSE- group of assessees, at least within the time
dimension (six months after assessment) covered by the
research and as measured by the instruments used.

If the research findings are replicated in other
organizations, supervisors who have pondered‘the impact of
assessment ratings on job performance and job satisfaction
may be reassured that the only area of concern seems to be
promotion satisfaction. Other than that, there is no

apparent impact six months after assessment.

LIST OF REFERENCES

LIST OF REFERENCES

Albrecht, Paul A.; Glaser, Edward M.; and Marks, John.

Bentz, V. Jan. "The Sears Experience in the Investigation,

"Validation of a Multiple-Assessment Procedure for
Managerial Personnel." Journal of Applied Psychology.
48 (December, 1964), 351-360. :8

 

Bernard M. "The Leaderless Group Discussion."
Psychological Bulletin, 51 (September, 1954),
465-492.

 

Description, and Predictation of Executive Behavior." 4
in Predicting Managerial Success. Edited by John A. g
Myers, Jr. Ann Arbor, Michigan: Foundation for a
Research on Human Behavior, 1968, pp. 59-152.

 

 

Bray, Douglas W. Issues in the Study of Talent. New York:

 

Columbia University Press, 1954}

. "The Management Progress Study." American

Psychologist, 19 (June, 1964), 419-420.

 

. "The Assessment Center: Opportunities for

Women." Personnel, 48 (September-October, 1971),

 

30-34.

and Campbell, Richard J. "Selection of Salesmen
by Means of an Assessment Center." Journal of Applied
Psychology, 52 (February, 1968), 36-41.

 

 

and Grant, Donald L. "The Assessment Center in
the Measurement of Potential for Business Manage-
ment." Psychology Monographs, 80 (1966), Whole No.
625.

 

and Moses, Joseph L. "Personnel Selection."

Annual Review of Psychology, 23 (1972), 545-576.

 

Byham, William C. "Assessment Centers for Spotting

Future Managers." Harvard Business Review, July-
August, 1970, pp. 150-160, 162-168, 17IlI72.

 

. "The Assessment Center as an Aid in Management
Development." Training and Develgpment Journal,
25 (December, 1971), 10422.

 

115

116

and Pentecost, Regina. "The Assessment Center:
Identifying Tomorrow's Managers." Personnel, 47
(September-October, 1970), 17-28.

 

Campbell, John P.; Dunnette, Marvin D.; Lawler, Edward E.,
III; and Weick,Karl E., Jr. Managerial Behavior,

Performance, and Effectiveness. New York: McGraw-
Hill, 1970.

 

Campbell, Richard J. and Bray, Douglas W. "Assessment
Centers: An Aid in Management Selection." Personnel
Administration, 30 (March-April, 1967), 6-13.

 

 

Carleton, Frederick 0. "Relationship Between Follow-Up
Evaluations and Information Developed in a Management
Assessment Center." Proceedings of American Psycho-
logical Association, 78 (1970), 565-566.

 

 

Cohen, Arthur R. "Some Implications of Self-Esteem for 1
Social Influence." in The Self in Social Interaction. E?
Edited by Chad Gordon and Kenneth J. Gergen. New
York: Wiley, 1968.

 

 

Cronbach, Lee J. Essentials in Psychological Testing.
New York: Harper & Row, 1960.

DiCostanzo, Frank and Andretta, Thomas. "The Supervisory
Assessment Center in the Internal Revenue Service."
Training and Development Journal, 24 (September,
1970), 12-15.

 

Dunnette, Marvin D. "Predictors of Executive Success."
in Measuring Executive Effectiveness. Edited by
Frederic R. Wickert and Dalton E. McFarland.

New York: Appleton-Century-Crofts, 1967, pp. 7-48.

. "The Assessment of Managerial Talent." in
Advances in Psychological Assessment, Vol. 2. Edited
by Paul McReynolds. Palo Alto, California: Science
& Behavior Books, 1971, pp. 79-108.

 

Eran, Mordechai. "Relationships Between Self-Perceived
Personality Traits and Job Attitudes in Middle
Management." Journal oprpplied Psychology, 50
(October, 1966) 424-430.

Eysenck, Hans J. Uses and Abuses of Psychology.
Baltimore: Penguin Books, 1953.

 

Ferguson, George A. Statistical Analysis in PsycholOgy
and Education. New York: McGraw-Hill, 1966.

 

 

117

Finkle, Robert B. and Jones, William S. Assessing
Corporate Talent: A Key to Managerial Manpower
PIanning. New York: Wiley, 1970}

 

Finley, Robert M., Jr. "Evaluation of Behavior Predic-
tions From Projective Tests Given in a Management
Assessment Center." Proceedings of American Psycho-
logical Association, 78 (1970), 567-568.

Garrett, Henry E. Statistics in Psychology_and Education.
New York: McKay, 1966.

Ghiselli, Edwin E. "Self-Description Inventory."
Berkeley: University of California, 1955 (Mimeo-
graphed).

. "Managerial Talent." American Psychologist,
18 (October, 1963), 631-642.

 

. Explorations in Managerial Talent. Pacific
Palisades, California: Goodyear Publishing, 1971.

Gordon, Chad and Gergen, Kenneth J. (Editors). The Self
in Social Interaction. New York: Wiley, 1968.

 

Grant, Donald L. and Bray, Douglas W. "Contributions of
the Interview to Assessment of Managerial Potential."
Journal of Applied Psychology, 53 (February, 1969),
24-34.

; Katovsky, Walter; and Bray, Douglas W. "Con-
tributions of Projective Techniques to Assessment
of Management Potential." Journal of Applied
Psychology, 51 (June, 1967), 226-232.

 

 

Guilford, J. P. Psyghometric Methods. New York: McGraw-
Hill, 1954.

Hardesty, D. L. and Jones, W. S. "Characteristics of
Judged High Potential Management Personnel -- The
Operations of an Industrial Center." Personnel
Psychology, 21 (Spring, 1968), 85-98.

 

 

Hinrichs, J. R. "Comparison of 'Real Life' Assessments
of Management Potential with Situational Exercises,
Paper-and—Pencil Ability Tests, and Personality
Inventories." Journal of Applied Psychology, 53
(October, 1969), 425-432.

Jaffee, Cabot L. "A Tridimensional Approach to Management
Selection." Personnel Journal, 46 (July-August,
1967), 453-455.

 

118

. Effective Management Selection: The Analysis
of Behavior by Simulation Techniques. Reading,
Massachusetts: Addison-Wesley, 197I.

; Bender, Joe; and Lynn, Calvert O. "The Assess-
ment Center Technique: A Validation Study."
Management of Personnel Quarterly, 9 (Fall, 1970),
9-14 0

Kay, Emanual; French, John R. P., Jr.; and Meyer,
Herbert H. "A Study of Threat and Participation in
an Industrial Performance Appraisal Program." General
Electric Company, Behavioral Research Service, May,
1962.

Kelly, B. Lowell. Assessment of Human Characteristics.
Belmont, California: Brooks/Cole, 1967.

and Fiske, Donald W. "The Prediction of Success
in the V. A. Training Program in Clinical Psychology."
American Psychologist, 5 (1950), 395—406.

 

The Prediction of Performance in Clinical
Psychology. Ann Arbor: UniVersity of’Michigan
Press, 1951.

 

 

Korman, Abraham K. "Self-Esteem Variable in Vocational
Choice." Journal of Applied ngchology, 50 (December,
1966), 479-486.

. "The Prediction of Managerial Performance: A
Review." Personnel Psychology, 21 (Autumn, 1968)a,
295-322.

. "Task Success, Task P0pularity and Self-Esteem
as Influence on Task Liking." Journal of Applied
Psychology, 52 (December, 1968)b, 484-490.

 

 

. "Toward a Hypothesis of WOrk Behavior."
Journal of Applied Psychology, 54 (February, 1970),
31-41 0

Kraut, Allen L. "A Hard Look at Management Assessment
Centers and Their Future." Personnel Journal, 51

 

and Scott, Grant J. "Validity of an Operational
Management Assessment Program." Journal of Applied
Psychology, 56 (April, 1972), 124-129.

 

 

Laurent, Harry. "Research on the Identification of
Management Potential." in Predicting Managerial

 

119

Success. Edited by John A. Myers, Jr. Ann Arbor,
Michigan: Foundation for Research on Human Be-
havior, 1968, pp. 1-34.

Lawshe, C. H. and Balma, Michael J. Principles of
Personnel Testing. New York: McGraw-Hill, 1966.

 

Lindzey, Cardner (Editor). Handbook of Social Psychology.
Cambridge, Massachusetts: Addison-Wes1ey, 1954.

LOpez, Felix M., Jr. Evaluating Executive Decision
Making: The In—basket Technique. AMA Research
Study No. 75. New York: American Management
Association, 1966.

 

McConnell, John H. "The Assessment Center in the
Smaller Company." Personnel, 46 (March-April,
1969), 40-46.

 

. "The Assessment Center: A Flexible Program
for Supervisors." Personnel, 48 (September-
October, 1971), 35-40.

 

Meyer, Herbert H. "An Evaluation of a Supervisory
Selection Program." Personnel ngchology, 9 (Winter,
1956), 499-513.

. "The Validity of the In-Basket Test as a
Measure of Managerial Performance." Personnel
Psychology, 23 (Autumn, 1970), 297-307.

 

Morris, Ben 8. "Officer Selection in the British Army,
1942-1945." Occupational Ppychology, 23 (October,
1949), 219-234.

Murray, Henry A. Exporations in Personality. New York:
Oxford University Press, 1938}

Myers, John A., Jr. (Editor). Predicting Managerial
Success. Ann Arbor, Michigan: Foundation for
Research on Human Behavior, 1968.

 

Nunnally, Jum C. PsyChometric Theory. New York:
McGraw-Hill, 1967.

 

OSS Assessment Staff. Assessment of Men. New York:
Rinehart, 1948. ‘

 

"PAR Effects Study." Pacific Telephone and Telegraph
Company, 1968.

120

"Personnel Assessment Program." Detroit, Michigan:
Michigan Bell Telephone Company, undated.

Randle, C. Wilson. How to Identify Promotable Execu-
tives." Harvard Business Review, May-June, 1956,
pp. 122—134.

Slevin, Dennis P. "The Assessment Center: Breakthrough
in Management Appraisal and DevelOpment." Personnel
Journal, 51 (April, 1972), 255-261.

 

Smith, Patricia Cain; Kendall, Lorne M.: and Hulin,
Charles L. The Measurementtof Satisfaction in WOrk
and Retirement. Chicago: Rand McNally, 1969.

 

Taft, Ronald. "Multiple Methods of Personality Assess-
ment." Psychological Bulletin, 56 (September, 1959),
333-352.

Thomson, Harvey A. "Comparison of Predictor Criterion
Judgments of Managerial Performance Using the
Multitrait-Multimethod Approach." Journal of
Applied Psychology, 54 (December, 1970), 496-502.

Wickert, Frederic R. and McFarland, Dalton E. (Editors).
Measuring Executive Effectiveness. New York:
Appleton-Century-Crofts, 1967.

 

Wikstrom, Walter S. "Assessing Managerial Talent."
The Conference Board Record, 4 (March, 1967),
9-44 0

 

Wollowick, Herbert B. and McNamara, W. J. "Relationship
of the Components of an Assessment Center to Manage-
ment Success." Journal of Applied Psychology, 53
(October, 1969), 348-352.

 

.8

fic".

APPENDICES

I.
4 . 4
N
.4-
IIIICI..|IIJ 1‘1. m

.0 . ..Ddlr

APPENDIX A

Ghiselli

Self-Description Inventory

 

121

 

 

msoumcmw umm oE ucmw mucH msomwmusoo
an S SH HHH SH S
umoco: ucmmmmHa manHuduwucm pmcHaumumc
m w mmuo mHSuNCI cow 2 quu w m comm
c u: H m Hm p 6 mm H m mH Hg 6 c m
mHanm o>HumHoouddm oHumHHmmu HSHOH
so we a mean mo a unumm u use comm :
u Hu on 6 mm Hx H; H «H p p H o
oHuwnumdamm memHuom ucmHonwm wcHum>mmuma
m mudm m moo cow: % 0 mac u Em
HS S mm H H Hm HH H mH H HS S
HmUHwOH pmmHOQ chx oHumwuocm
u so 0 c Sun m oumum o a How u
x H mm p c H H m cm nHH p «H H w n s
mDOHucoHomcoo Homunwsocu pmuuHaldpmnm mecmHum
pmHHouucooleom ano uuon m>Hucw>cH
m>HumcmeaH mm muwoch mH pouomummc: HH m>HumHodooo m
on H> o So u u : wousomwu woouo u
p HH. H om HH 5 wH H m 0H s n N
pmeHcch m>Hmwmuwoua Homcde wchcmumumcco
m m on own mo uumud ummuom
Hp HH mm x m NH H H m Hp H
mHnHchQmmu wHSCOHuommwm mDOHuumoch oHnmdmo

.som mmbHuummp umoE xchu so» pucB moo ocu x0050 onwn mcuo3 mo mHHma mnu mo comm cH

.muHmd wcHBOHHow mnu mo comm cH ppoB mco mchmn :x: cm mumHm .cmo :0» mm mHummcon com mHouwH
Isuom mm memuso> mnHHomop ou mug om .mum3mcm mocha no uLwHH o: mum ouwcH .wHomH:o% mnHHommc 90% so:
mom cu pom mmwmmom 30> o>mHHmb so» muHmHu wnu mo muouoHd m chuno ou mH >H0uco>cH chu mo uncouod och

 

VMOBzm>ZH ZOHHmHmUmMQIMHmm

122

ucwHumdsH

mHnHchQSwHuH

mHnmuHoxo

%Hm
Howuchmu
prmccmdmpco
cmeQmuso
pmemHummch
o>Hmmwuwwm
Hmocho
wchqulemm
pmuummnupumc
Hmuqu

SSSHSSHSS

oHumHEHmmmd

pwumCOHcHdo

‘mht.’ 11. . blind

so

mo

No

He

oo

mm

mm

mm

.30»

m>HmmHEnsw om acmumHOucH we pHou
ammSH mao>umc chonpsum
% pm wow m and % 005
H c Hp mm 5 H8 no u
pochHnlmHuumu mHnmmwcwco pouommwm
cmumucmolmHmm an ucopcodwp be wamemmlemm
opsu hammuv thcwHumco
m mm o mu up maom muumsv
c HwH mm Ha H H ms H
xmma mmcou muaumaaH
mo c m dEoo % coco wcouum mm
H H H mm a as p ;
oHnHuomuuch m>Hmcmwop HmaOHuoEm
w> mm>w woo o>Hu ucmwouum
H Hm H . w ms
unopCOQmmp mHnmumcs AmHo:
mo um uowm zwcHum mmm omu
H H H on . Ne Hx
UHumnumdm soHHmnm maoHanamca
mH oo o How a mum
r H m as HH w H Hq H
mmmHmHmo pwuHmucoo %nm

mwbHHommc ummmH xcHLu 30% who: moo mnu sumac 30H0n mpuo3 mo HHmd 0:» mo comm

oq

mm

mm

mm

on

mm

«m

mm

CH

APPENDIX B

Job Descriptive Inventory (JDI)

123

unmaanHdaooom Ho wwcwm mm>HU ucmmmem

mmmecm uom
oHdaHm 1|III1 pouomdmmm
wcHumuuwsum IIIII1 m>Hummuu

umow know so .IIIII coco .IIIII
onwcmHHmno IIIII1 wcHuom
H3818: Ill wcHaHSHSSS
waomwuHH IIIIII wcHusom

Hummus IIIII. wcHumcHommm 1III11

 

Ill

.EmuH mnu mchmb :w: m mUMHd .aouH mcu u30bm opHump uoccmo 30% HH .6

.SmuH on»
oprmn :2: no mode .xuoa usom mo uowdmm umHsoHuumd m mbHHommp uo: wwop EmuH mnu HH .n

 

.amuH ocu opHmwb RM: m wode .xuos Hsom mo uomdmm umHsoHuHmd m mobHuommp amuH osu NH .m

"Bonn mamuH ecu pom

 

xuoz .H COHuomm

124

 

.. I'lJWV

1.lr

cmcmmc coxa pcsou< pwumaemu onso

HNSH .Illl1 nwsocw mmH>uwa36 u.cwmon
sac >5 so 05 mm>me .IIIII mumcnoulda IIIII.
HommHHkucH IIIII1 HmHucmaHmcH 1IIIII
pmm IIIII. stuoma
HHSB 20h mzocM .IIIII xuoa coow mmmHmum
auonnoum .IIIII ouHHoaSH
onmood< .IIIII mmmde cu puma .IIIII
ccmum H muons ma mHHmH .IIIII moH>cm he mxm< .lllll

 

.EwuH mzu mpHmmn :m: m womHa .EmuH msu usonm mpHomc uoccmo 50% NH .0

 

.EmHH mLu mchmn :2: cm mode .uomH>umdsm poo» mo uomdmm cm mnHHomoc uoc mmoc aouH msu HH .n
.EmuH mzu mpHmmn :w: m mode .HomH>deom uso> Ho Homawm HmHSUHuHmQ m mmnHHommp EmuH onu HH .m

HBOHmn meuH mnu pom

 

conH>Hmdom .HH GOHHUmm

125

.1551.
4
n8

w!

coHuoaoud How mocmno poow SHHHmm
mcoHuosoud umHsmmm

chHuosoud ucmsvaHcH

moHHom :OHuoaoud HHmmc:
COHuoaoud HoH mocmno coco

SSH Scm1emoe

muHHHno co coHuoaoum

cmuHBHH umsaoaom HuHcsuHOQQO

ucoamucm>pm now AuHcauHOQQO pooo

 

.EmuH mnu mpHmwn :w: m mode .emuH may uoonm opHomp uoccmo 30% HH .0

oEQUH USU QUHmmfl. :z:
cm mode .Eoumxm coHuoEoud Hack Ho uoodmm umHsoHuumd m wnHHummp uoc mwOp EmuH wnu HH .3

 

.EouH mcu
mpHmwn z»: m wode .Emumzm COHuoEoud H50» Ho uomdmm HmHsoHuHmd m mmbHHommp EmuH wnu HH .m

 

"SOHmn maouH who pom

 

mGOHuoeoum .HHH COHuoom

126

umoa cu chum .IIIII mmHamcm oxme cu zmmm
HomOH .IIIII ucmeHHmucH
mumoumucH Bouumz IIIII. ummm IIIII1
m>Huo< .IIIII mHnHchQSmm .lllll
>0m>HHd oz .lllll pHQSum .lllll
BadmmdecD .lllll mDOHana<
HNmH IIIII1 30Hm
uHmEm IIIII. wcHHom
Luna cow HHmH IIIII1 wcHumHsaHum

 

.EmuH mnu mpHmmn :w: m mode .BmuH wnu usonm mpHump uoccmo :0» HH .0

 

.EmuH on» mpHmmn :z: cm mode .mumxuoanoo Hoom Ho uomamm cm mnHuowmp uoc moOp EmuH mnu HH .n
.EMDH och mchmb :w: m mode .mHoxH031ou use» Ho Homdmm HmHsoHuHmd m mmpHHommp amuH who HH .m

"BOHon mEouH mnu pom

 

mumxuozlou .>H COHuowm

 

APPENDIX C

Performance Self-Rating Form

 

127

SELF-RATING FORM FOR:

 

(Enter your name here)

INSTRUCTIONS

This form serves two purposes. First, it will introduce you to the
skills which are evaluated in the Assessment Center. You will rate
yourself on these skills with regard to your own performance back on
your job. As you read through the skills and their definitions, you
should be aware that they have been identified by Division management
as critical to managerial success.

The second function of this rating form is purely research. The Divi-
sion, in connection with Personnel and Organization Staff, is conduct—
ing an evaluation study of the Center to determine how much of the
information collected at the Center could be obtained from pe0ple who
merely observe you on your job. That obviously includes yourself as
an observer. This week your supervisor and a few of your own co-
workers will also rate you on a form almost identical to this one.
When all the ratings are collected on approximately 60 Center partici—
pants this year, the different sources of information will be compared.

NONE OF THE RATINGS GIVEN BY YOUR SUPERVISOR, YOUR COWORKERS, OR YOUR-
SELF WILL BE A PART OF YOUR PERMANENT RECORD. THEY WILL ONLY ASSIST
US IN EVALUATING THE CENTER.

Follow these steps in rating yourself:
1. Read the definition provided for each skill.

2. Read the definition a second time and think about the meaning
we are trying to convey.

3. Think of your own job requirements and decide whether you
have had enough opportunities to test your performance on
the particular skill in order to give yourself a rating.

4. If you feel that your job has not given you ample Oppor-
tunities to test and measure the skill, place a check in
the box to indicate this.

5. If you do have a good idea of your job performance on the
skill, rate yourself by circling a number along the scale.

6. Briefly give an example of your typical behavior which led
you to give the particular rating. Be specific.

FOR RESEARCH PURPOSES ONLY

Skill:

 

1.2(3

or a task without creating hostility?
«1 Lb people look to you for direction?
0 Do people respond to you as a leader, not just a boss?

0 Are you able to get people to follow you in the solution

R1tinn: [::] My Job has not given me the opportunity to t>st and measure this

ability.

(If you have not checked the box above, circle your rating of yourself.)

1 P 3
10W BELOW
Although 1 AVERAGE
haw: had the I have
opu1rtunity, shown
I have shown little of
“our of‘ this this Skill
skill in my in my work
work.
l‘lxumple:

Skill: Lhcision Making, 0

 

O

O

4 5 6

SATISFACTORY

I have shown an
adequate amount
of this skill in
my work.

Are you able to seek out and evaluate pertinent

7 8

ABOVE

AVERAGE

I have shown
an above aver-
age amount of
this skill in
my work.

facts, and make sound Judgments?

Are you able to put elements of your work into

meaningful priorities?

9

EXCEPTIONAL

I have con-
sistently
shown as much
of this skill
as could be i
expected of
a person in

my Job.

 

Are you capable of diacrbminating between relevant

and irrelevant facts?

Hating: [::] My Job has not given me the opportunity to test and measure this

ability.

(If you have not checked the box above, circle your rating of yourself.)

1 2 3
low BELOW
AVERAGE

Example:

A 5 6

SATISFACTORY

(See above for explanation)

7 8

ABOVE
AVERAGE

9

EXCEPTIONAL

;H.ll l: le'l.:iv‘

1.259

:u.u- <1 1x1you recognize when a decision is necessary

immediately and respond on the basis of the
information available, rather than putting off
the decision?

It1thug: [::] My job has not given me the opportunity to test and measure this

(If you hWVV not

1 2.)

10W
Al thmn'll I
hILVv- lflld tin“
opportunity.
1 llrlw' 111101131"
noun of this
skill in my
work.

l'i‘fituuplt':

ability.

checked the lox.abovc, circle your rating or yourself.)

3

BELOW
AVERAGE

I have
shown
little of
this skill
in my work

A

O

Skill: Organizational Ability o

 

O

O

O

5 6

SATISFACTORY

I have shown an
adequate amount
of this skill in
my work.

7

ABOVE

AVERAGE

I have shown
an above aver
age amount of
this skill in
my work.

Do you show the ability to plan and
organize the work of others?
Do you delegate, when given the opportunity,

and establish administrative controls?

8

9

EXCEPrIONAL

I have con-
sistently
shown as much
of this skill
as could be
expected of

a person in
my Job.

Do you set up schedules so that deadlines can

be met?

Do you take into account the long range effects

of your plans.

Hating: [::] My job has not given me the opportunity to test and measure this

(If you have not checked the box above, circle your rating of yourself.)

1 ?

IOW

Example:

ability.

3

BELOW
AVERAGE

A

(See above for explanation)

S 6

SATISFACTORY

7

ABOVE
AVERAGE

8

9

EXCEPTIONAL

130

ilkill: Initirnlvv 0 D11 you actively uttvmpt to influence people or events
or do you passively go along with the group?
n Aw you a self-starter? Are you the one to get the
lr'Lll rolling?

Hating: [::] My Job has not given me the opportunity to test and measure this
:111111 ty .

(ll' yHH haw. 11111. checked the box above, circle your rating of yourself.)

1 'r' «i ll- 5 (J 7 8 9
lOW BELOW SNPISFACTORY ABOVE EXECPTIONAL
A_l.1l;(1:1.-ih 1 AVERAGE I have shown an AVERAGE I have con-
thu had the I have adequate amount I have shown sistently
opportunity, shown of this skill in an above aver- shown as much
I h:1v<- shown little of my work. age amount of of this skill
none 01' this this skill this skill in as could be
skill in my in my work. my work. expected of
work. a person in
my job.
{'1 rumpltf- :

Skill: Rcuponnc to Changing Conditions 0 When faced with changing condi-
tions or new information, are
you able to generate alternative
actions or directions?

o to you become flustcred when n
sudden change occurs, or are you
able to deal with the situation
successfully?

hntihr: [::] My Job has not given me the opportunity to test and measure th1s
ability.

(1r y1n have not checked the box above, circle your rating of yourself.)

1 R 3 4 5 6 T 8 9
I ()W BELOW SATISFACTO HY MOVE EXCEPTIONAL
AVERAGE AVERAGE

(See above for explanation)

I",:~:r1mp L":

1231

Skill: Hi cwrnmvnt 0 Are you able to accurately perceive the strengths and
weaknesses of others with whom you must interact?
o Are you able to size up the effectiveness of coworkers?

Uiillhn [::] My job has not given me the opportunity to test and measure this
‘lll’l iity .

(If you hnvv not checked the box above, circle your rating of yourself.)

1 3 h 5 6 T - 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although ] AVERAGE I have shown an AVERAGE I have con-
hnvo had the I have adequate amount I have shown sistently
opportunity, shown of this skill in an above aver- shown as much
I have shown little of my work. age amount of of this skill
none of this this skill this skill in as could be
skill in my in my work. my work. expected of
work. a person in

my Job.
l'I:<:meJ.L‘:

Skill: Oral Communications 0 Can you clearly and effectively present your
point of view in a meeting situation?
0 Do people understand what you are talking about
when engaged in a conversation with you?
0 Do you sound confident and well organized?

Hutinfl: [::] My Job has not given me the opportunity to test and measure this

ability.
(Ir you have not checked the box above, circle your rating of yourself.)
1 2 3 h 5 6 7 8 9
low lflHOW SATISFACTORY ABOVE EXCEPTIONAL
AVERAGE AVERAGE

(See above for explanation)

:»:;um_~i« 1

Skill:

Written Communications

 

Rating:

(If you have not checked the box above, circle your rating of yourself.)

1 P

10W
Although I
haw had the
opportunity,
I have shown
none of this
skill in my
work.

Example:

Ukill:

huiinn:

D lrgation

3

BELOW
AVERAGE

I have
shown
little of
this skill
in my work.

A

132

Can you present written information in a

logical and understandable form?
Do you highlight mador points?
Is your grammar adequate?

5 6

SATISFACTORY

I have shown an
adequate amount
of this skill in
my work.

7 8

ABOVE

AVERAGE

I have shown
an above aver-
age amount of
this skill in
my work.

0 Do you delegate appropriately to others

when given the opportunity?

[::] My Job has not given me the opportunity to test and measure

ability.

(It you have not checked the box above, circle your rating of yourself.)

1 :3 5 h 5 6 7 8
LOW BELOW SATISFACTORY ABOVE
Although i AVERAGE I have shown an AVERAGE
have had the I have adequate amount I have shown
opportunity, shown of this skill in an above aver-
I have shown little of my work. age amount of
none of this this skill this skill in

skill in my
Wk)lk c

in my work.

my work.

[::] My Job has not given me the opportunity to test and measure this
ability.

9

EXECPTIONAL

I have con-
sistently
shown as much
of this skill
as could be
expected of

a person in
my Job.

 

”'1'

this

9
EXCEPTIONAL
I have con-
sistently
shown as much
of this skill
as could be
expected of
a person in
my Job.

 

APPENDIX D

Supervisor/Co-worker

Performance Rating Form
(Before Assessment)

 

This form is to be used to rate the performance of

133

INSTRUCTIONS

 

 

Your relationship to this individual (Check one):

Have you ever attended the Assessment Center?

His supervisor

in

His Coworker

No

.‘h II

Yes, as an observer.

u

’-

 

]? -_V

Yes, as a participant.

This rating form is part of a research study to assist the Division in
evaluating the Assessment Center. The ratings which you are asked to
give the individual named above can neither help nor hurt him; your
ratings will not become a part of his permanent record.

On the following pages you will find a number of job skills which are
important for management personnel.

1.

2.

Read the definition provided for each skill.

Read the definition a second time and think about the
meaning we are trying to convey.

Think of your eXperiences with the individual being rated
and the requirements of his job and decide whether you
have seen enough of his performance to enable you to give
him a rating.

If you feel that you have not observed the individual
with regard to a specific skill, place a check in the
box to indicate this.

If you have observed him sufficiently, give him a rating
by circling a number along the scale.

Briefly give an example of his typical behavior which led
you to give him the particular rating. Be Specific.

lEi4

£23 RESEARCH PURPOSES ONLY

Skill: Leadership 0 Is the individual able to get people to follow him
in the solution of a task without creating hostility?
0 Do people look to him for direction?
0 Do people respond to him as a leader, not Just a boss?

Rating: [::] I have not observed him enough to give a rating on this skill.

(If you have not checked the box above, circle your rating of this individual.)

1 2 3 h 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown amount of this an above aver- shown as much
he has shown little of skill in his age amount of of this skill
none of this this skill work. this skill in as could be
skill in his in his work. his work. expected of
work. a person in
his Job.
Example:

Skill: Decision Making 0

 

O

0

Rating: [::] I have not

Is the individual able to seek out and evaluate
pertinent facts, and make sound Judgments?

Is he able to put elements of his work into
meaningful priorities?

Is he capable of discriminating between relevant
and irrelevant facts?

observed him enough to give a rating on this skill.

(If you have not checked the box above, circle your rating of this individual.)

1 2 3
LOW BELOW
AVERAGE

Example:

A 5 6 7 8 9
SATISFACTORY ABOVE EXCEPTIONAL
AVERAGE

(See above for explanation)

 

135

Skill: Decisiveness 0 Does he recognize when a decision is necessary
immediately and respond on the basis of the
information available, rather than putting off
the decision?

Rating: B I have not observed him enough to give a rating on this skill.

If!

 

(If you have not checked the box above, circle your rating of this individual.)
1 2 3 u 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown amount of this an above aver- shown as much
he has shown little of skill in his age amount of of this skill
none of this this skill work. this skill in as could be
skill in his in his work. his work. expected of E5
work. a person in
his Job.
Example:

Skill: Organizational 0 Does he show the ability to plan and organize
Abilit the work of others?
'—__-_JL 0 Does he delegate, when given the opportunity,
and establish administrative controls?
0 Does he set up schedules so that deadlines can
be met?
0 Does he take into account the long range effects
of his plans?

Rating: [::] I have not observed him enough to give a rating on this skill.

(If you have not checked the box above, circle your rating of this individual.)

1 2 3 1+ 5 6 7 8 9
Low BELOW SATISFACTORY ABOVE EXCEPTIONAL
AVERAGE AVERAGE

See above for explanation)

Example:

136

Skill: Initiative 0 Does he actively attempt to influence people or

events or does he passively go along with the
group?

0 Is he a self-starter? Is he the one to get the
ball rolling?

Rating: [::] I have not observed him enough to give a rating on this skill.

(If you have not checked the box above, circle your rating of this individual.)

1

LOW
Although he
has had the
opportunity,
he has Shown
none of this
skill in his
work.

Example:

2 3 1+ 5 6 7 8 9
BELOW SATISFACTORY ABOVE EXCEPTIONAL
AVERAGE He has shown AVERAGE He has con-
He has an adequate He has shown sistently
shown amount of this an above aver- shown as much
little of skill in his age amount of of this skill
this skill work. this skill in as could be
in his work. his work. expected of
a person in
his Job.

Skill: Response to Changing Conditions 0 When faced with changing con-

Rating:

ditions or new information, is
he able to generate alternative
actions or directions?

0 Does he become flustered when a
sudden change occurs, or is he
able to deal with the situation
successfully?

D I have not observed him enough to give a rating on this skill.

(If you have not checked the box above, circle your rating of this individual.)

1 2

LOW

Example:

3 1* S 6 7 8 9
BELOW SATISFACTORY ABOVE EXCEPTIONAL
AVERAGE AVERAGE

(See above for explanation)

137

Skill: Discernment o Is he able to accurately perceive the strengths
and weaknesses of others with whom he must interact?
o Is he able to size up the effectiveness of coworkers?

 

 

Rating: I have not observed him enough to give a rating on this skill.

 

(If you have not checked the box above, circle your rating of this individual.)

1 2 3 u 5 6 7 8 9

LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown amount of this an above aver- Shown as much
he has shown little of skill in his age amount of of this skill
none of this this skill work. this skill in as could be
skill in his in his work. his work. expected of
work. a person in

his Job.
Example:

Skill: Oral Communications 0 Can he clearly and effectively present his
point of View in a meeting situation?
0 Do people understand what he is talking about
when engaged in a conversation with him?
0 Does he sound confident and well organized?

Rating: 1 I I have not observed him enough to give a rating on this skill.

 

(If you have not checked the box above, circle your rating of this individual.)

1 2 3 1+ 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE ' EXCEPTIONAL
AVERAGE AVERAGE

(See above for explanation)

Example:

138

Skill: Written Communications 0 Can he present written information in a

logical and understandable form?
0 Does he highlight major points?
0 Is his grammar adequate?

Rating: I:

.If you have not checked the box above, circle your rating of this individual.)

I have not observed him enough to give a rating on this skill.

1 2 3 1+ 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown amount of this an above aver- shown as much
he has shown little of skill in his age amount of as could be
none of this this skill work. this skill in expected of
skill in his in his work. his work. a person in
work. his Job.

Example:
Skill: Delegation 0 Does he delegate appropriately to others

when given the opportunity?

Rating: l:::] I have not observed him enough to give a rating on this skill.
{If you have not checked the box above, circle your rating of this individual.)
1 2 3 h 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCECEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown little amount of this an above aver- shown as much
he has shown of this skill in his age amount of as could be
none of this skill in work. this skill in expected of
skill in his his work. his work. a person in
work. his Job.

Example:

 

APPENDIX E

Supervisor Performance Rating Form
(After Assessment)

 

139

This form is to be used to rate the job performance of

 

This individual ___*__ does-- _______dons not-—report to me now.
Please circle the appropriate number to indicate your rating on each skill. Check the box
if you feel that you cannot give a rating.

Have you seen any overall change in this subordinate's Job performance since he attended
the Assessment Center?

Great deterioration Slight Some Great
W Maxims clans: W Wei.
1 2 3 1+ 5

Skill: Le e hi 0 Is the individual able to get people to follow him.in the solution
of a task without creating hostility?

0 Do people look to him for direction?
0 Do peOple respond to him as a leader, not just a boss?

Rating: [::] I have not observed him.enough to give a rating on this skill.
(If you have not checked the box above, circle your rating of this individual )

l 2 3 ’+ 5 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTIONAL
Although he AVERAGE He has shown AVERAGE He has con-
has had the He has an adequate He has shown sistently
opportunity, shown amount of this an above aver- shown as much
he has shown little of skill in his age amount of of this skill
none of this this skill work. this skill in as could be
skill in his in his work. his work. expected of
work. a person in
his job.

Skill: Do i ion Makin o Is the individual able to seek out and evaluate pertinent facts,
and make sound judgments?
o Is be able to put elements of his work into meaningful priorities?
0 Is be capable of discriminating between relevant and irrelevant

 

facts?
Rating: [:1 I have not observed him enough to give a rating on this skill.
l .1 3 ’4 5 6 7 8 9
Skill: Dnyigiyvnupg n ”HUS ho recognizn when a decision is necessary immediately and

respond on the basis of the information available, rather than
putting off the decision?

Rating: [::] l haVu not observed him enough to give a rating on this Skill.

1 R 3 h S 6 7 8 9

140

Skill: W 0 Does he show the acuity to plan and organize the work or Others:
ANNEX 0 Does he delegate, when given the Opportunity, and establish ad-
ministrative controls?
0 Does he set up schedules so that deadlines can be met?
0 Does he take into account the long range effects of his plans?

Rating: [3 I have not observed him enough to give a rating on this skill.

1 2 3 1+ s 6 7 8 9
LOW BELOW SATISFACTORY ABOVE EXCEPTICNAL
- AVERAGE AVERAGE
Skill: W 0 Does he delegate apprOpriately to others when given the opportunity?

Rating: B I have not observed him enough to give a rating on this skill.
1 2 3 u s 6 7 8 9
Skill: m 0 Does he actively attempt to influence peeple or events or does he
passively go along with the group?
0 Is he a self-starter? Is he the one to get the ball rolling?

Rating: [:1 I have not observed him enough to give a rating on this skill.

1 2 3 l: 5 6 7 8 9
Skill: Wing c When faced with changing conditions or new information, is
W he able to generate alternative actions or directions?

0 Does he become flustered when a sudden change occurs, or
is he able to deal with the situation successfully?

Rating: E] I have not observed him enough to give a rating on this skill.
1 2 3 1+ 5 6 7 8 9
Skill: W o Is he able to accurately perceive the strengths and weaknesses of
others with whom he must interact?
o Is he able to size up the effectiveness of coworkers?
Rating: B I have not observed him enough to give a rating on this skill.
1 2 3 1+ 5 6 7 8 9
Skill: gm; 0 Can he clearly and effectively present his point of view in a

C 10 meeting situation?

0 Do people understand what he is talking about when engaged in
a conversation with him?
0 Does he sound confident and well organized?

 

Rating: B l have not observed him enough to give a rating on this skill.
1 2 3 u 5 6 7 8 9
Skill: Written 0 Can he present written information in a logical and understand-
C m tio able form?
0 Does he highlight major points?
0 Is his grammar adequate?

Rating: [:1 I have not observed him enough to give a rating on this skill.

1 2 3 1+ 5 6 7 8 9

 

Imm'mltmm'

IN

9
3
5
7
7
7
1
3
0
3
9

R"
I"
Em
"I

'1’

 

MW