HomeMy WebLinkAbout1982-11-30 Info Packetr
i
REPRINT
What is an assessment center?
The answer is not simple, and it depends on who you ask!
Cabot L. Jaffee and Joseph T. Sefcik, Jr.
ssessment centers are gaining in popular-
ity more.rapidly than perhaps any other
human resources tool and Ihev are
by no means a standardized tech-
nology. More and more people
are having their lives touched by the method-
ology and often the methodology is different,
dkhough called an assessment center. This is
both exciting and disturbing.
For many years, the unfilled assessment center
seminar rooms demonstrated the hostility cen.
tered around the protection of the more Tradi-
tional selection procedures and the overwhelm.
Ing feelings concerning the unrealistic time de.
mands and the general cost of the process. The
passage of time and the government have taken
care of many of these issues and the rooms are
now filled with people wanting to hear about
assessment centers. Traditional testing proce.
dures have fallen in the wake of validity require.
ments. The seemingly unrealistic time and money
costs have paled in comparison to the costs of
lawsuits, the payment of reparations and the
heightened concern for the welfare and fair
treatment of every individual.
They can be used any place in which information
is lacking, based on the person's past behavior, or
In any critical skill areas. Assessment centers
would be equally useful for entry-level police of.
ficers, teachers, salespeople, or a variety of differ.
ent kinds of managers. The major determinant of
usefulness is economics. An analysis of how many
people would be moved into the target position
over a defined period of time and the cost of an
error In the selection process are the major
economic barometers.
Any person who has read even superficially In
the area has seen the statement: "An assessment
center is not a place but rather a process." This
statement needed to be made because the first
assessment centers were places rather than a
standardized process. The use of work samples
or their combination with other types of instru.
ments have been the critical characteristics of
what is now accepted as the assessment center
method. It is from this point that the definition of
an assessment center needs to be addressed.
What is an Assessment Center?
It is a process in which individuals have an oppor-
tunity to pardcipale In a series of situations which
resemble what they might be called upon to do in
the real world. They are tested by situational or
simulation exercises and multiple -trained asses.
sors process Information In a fair and impartial
manner.
Where can Assessment Centers
be Used?
The questions stlll remain: what is described by
the term "assessment center" and why and how is
ft different from other processes used In the selec-
tion and development of people7 our views are
presented here with apologies to professional col.
leagues who honestly and with cause disagree,
and with no apologies to the ever-increasing
number of people who are now "selling" this
product with no more understanding of the con-
cept than the first-time user.
t
MICROFILMED BY
JORM MICROLAB
I DEDAR RAPIDS • DES M0196
i
Dr. Cabot L Jaffee is
President and Founder
of Assessment Designs,
Inc. He has designed
assessment centers for
the government and
hundreds of
organizations. Jaffee
holds a BA in
Psychology from New
York University, an MA
from Columbia
University and a PhD in
psychology from Florida
State University.
Joseph T. Sefclk Jr. is
Manager of Assessment
Center Services for
Assessment Designs
Inc He holds a BS in
psychology from Florida
Southern College with
on emphasis in business
and an MS in
psychology from the
University of Central
Florida with an
emphasis ht Industrial
psychology.
a.0,3S
J
WN
r
ASSESSMENT
CENTERS
C
How Do You Know What People
Will Be Called Upon To Do in
the Real World?
The only way to know is by studying the position.
This can be accomplished in a number of ways. all
of which include some type of job analysis. The
job analysis defines critical tasks and skills neces-
sary for success in the job.
What Makes a Good Situational
or Simulation Exercise?
It is determined by the degree to which there is
overlap or agreement between the test situation
and the real situation. The more the test situation
reflects the job demands and calls for the exhibi-
tion of skills that will be required on the job, the
better the ability of the tests to predict later job
behavior.
Why Do Most Assessment Centers
Measure the Same Skills?
They don't! Even though most assessmentcen-
ter's design had "organizing and planning" as a
necessary skill, every assessment center has not
been the same. The description of the skill in
terms of breaking a task down to make work
easier is not the real definition of the skill. The real
definition is the situational exercise used to mea-
sure the skill. For example, a first-line supervisor
organizes and plans as does a chief executive
officer. The situationally relevant lest to assess
planning for one would not likely be relevant for
the other. They are all called Organizing and
Planning because our language is not precise
enough to differentiate, but they are probably
totally different operations. The skills are not the
same, therefore. because the measures need to
be totally different.
How Many Skills Can You Measure?
The more skills you have, the greater chance of
overlap between them and the greater chance
that one piece of behavior will contribute signifi-
cantly to many skills. The more this may happen,
the more unreliable the assessment center. It is my
experience that approximately 10 skills, depend-
ing upon the complexity of the job, may be reabs-
tic — 20 or more are not
Should You Use Personality Tests or
Intelligence Tests in Order to Learn
More About a Person?
When you do this, you run the risk of misusing the
information from those tests. For example, if
somebody organizes and plans extremely well in
an exercise which resembles the job, what have
you learned if they appear unorganized on a per-
sonality test? If someone perceives the critical
elements of a business-related problem, why is it
important to know whether or not they can un-
derstand the correct answer to the analogy apples
are to oranges as pears are to bananas, cherries,
or plums? The very constructs measured by per-
sonality or intelligence tests can be measured by
situational tests in a much more relevant way.
What About The Personal Interview?
This falls prey to the problem of interpreting the
data. Intellectual responses in an interview setting
do not necessarily relate to a person's ability to
demonstrate the skill in other behavioral ways. A
person may do quite well at answering questions
concerning what they might do during critical
periods in a board room. but be totally unable to
effect the behaviors in the applied. job-related
setting. In addition, the information appears
less reliable and more subject to interpretation,
which is what the assessment center attempts to
recognize.
Can You Fake Performance
at an Assessment Center?
Can You Play a Role?
If someone were given a try -out for a basketball
team and they were able to score many points,
there would be very little question as to whether
the behavior was "real" or not. Whether or not
they choose to use these skills, over time, be-
comes a question of their interest, motivation or
opportunity, more than their skill level. Assess-
ment centers do not measure motivation — they
measure skill level and potential. But the know-
ledge of skill level can and should be of critical
importance in job performance. People who do
things well are likely to receive rewards from
doing these things and are therefore more likely to
perform In an effective manner. It then becomes
the responsibility of the organization to ensure the
availability of rewards for demonstration of effec-
tive performance.
Perhaps, in many ways, this has been the prob-
lem of the "disadvantaged". It is very difficult to
demonstrate a relationship between what they do
and what will happen to them. This fact is perhaps
a major reason why traditional tests of motivation
and measurements of job performance have not
appeared similar for advantaged and disadvan-
taged populations (Myart vs. Motorola, 1964;
Lopez,1966; Bartlett and O'Leary, 1969; Hicks vs.
Crown Zellerbach, 1970; and Griggs vs. Duke
Power Company, 1971). The evidence from the
assessment center results indicate that males and
females, and minority and majority candidates
perform more similarly to each other than theydo
on pencil -and -paper test. (Moses and Boehm,
1975; Huck and Bray, 1976; Frank and Drucker,
1977; Schmitt and Hill, 1977 ) This maybe due to a
measure of skill that Is present without having it
confounded by the motivation of the individual to
perform over a period of time. So perhaps moti-
vation decreases due to a lack of opportunity for
rewards over time in some organizations and the
shift level measured at the assessment center may
MICROFILMED BY +�
JORM MICROLAB
CEDAR RAPIDS •DCS h101AE5
ao3s
1
P1
v
r
be a clearer picture of potential than actual job
performance.
Why Have Some Studies Shown
Assessment Centers Predict
Better for Two Levels Above
a Candidate's Position Than One?
Because, insonte cases, these assessment centers
have probably been better simulations of the job
two levels above than one level above. This may
mean that more careful job analysis would pro-
vide better information for the development
of assessment center instruments that would pre-
dict performance one level above a candidate's
position.
How Do You Know Assessment '
Centers Are Valid?
Content validity is specified by the new guidelines
on employee selection (Federal Register 1978) as
an acceptable form of validity and it is to this type
of validity particularly that the assessment center
methodology lends itself. This type of validity is
inherent in the assessment center process when
the situational exercises are developed based upon
a thorough job analysis. Additionally, assessment
centers have demonstrated a strong positive vela.
tionship between ratings and future performance
on the job. This predictive validity has
been observed on many occasions; most no.
tably in the AT&T Management Prugress Study
(Bray and Grant. 1966), and by other research-
ers (Bray and Campbell. 1968: Jaffee. Bender
,end Calvert. 1970; Kraut and Scutt, 1972: Muck.
1973: Moses. 1973; Worbois. 1975; and Hinrichs.
1978).
How Well are Assessment Centers
Accepted?
Accepted by whom? If assessment centers are
well-run.and consider the needs of the people in
the organization, management, assessors and
candidates like them very much. Management
likes thein because they provide valuable data
upon which selection decisions can be made or
training programs developed. Assessors feel that
they are better managers as a result of having
participated in the process. And candidates like
them because they see them as being the basis of
a fair evaluation.
However, when assessment centers are used
by the organization as a "kiss of death" or an
irreversible labelling process, it can build up feel-
ings of discontent among those who may not
immediately obtain a desired position. So the
'kiss of death" has to do wRth how the data are
used rather than whether or not the data are
yuod. This says that assessment centers must be
entered into and developed thoughtfully so that
they provide the basis for effective human re.
sources utilization.
L�
Will Assessment Centers Be
Replaced by Some Other Process?
The major tenets of the assessment center are
built on a logical foundation. The ideas of doing a
job analysis, designing work samples to measure
those characteristics deemed critical and the train-
ing of indviduals to make judgments on those
critical skills are unassailable. To what extent any
given assessment center meets these steps is quite
another issue. However, the use of line managers
to ensure acceptance of the program and the
conscientious application of the critical steps in
the process should guarantee continued accep-
tance of the program by all concerned parties.
The only replacement for assessment centers,
then, is more effective development of simula.
tions that better represent the environment and
ways of teaching people to be more accurate
evaluators of critical behavior. Computers pro-
vide a good deal of hope for the more accurate
simulation of environments.
Do Assessment Centers Need to
be Designed by Psychologists?
They need to be designed by people who know
not only how to direct or perform the steps of job
analysis, exercise development and staff training,
but also individuals who can ensure the process
fits within an existing system which may consider
issues like career palhing, development, counsel.
ling and selection. Obviously, these people do not
need to be psychologists; they do need experi-
ence with Human Resource Management Sys-
tems. However, sensitivity to administrative and
possibly research method issues as they relate to
equal opportunities and fairness in test administra-
tion are important for the success of assessment
center programs. Individuals with prescribed
training in research design, like psychologists.
might be helpful. But certainlyissues such as
exercise order, time limits, daily schedules,
exercise/candidate level match, etc. are things
of which these people should be aware
(Cohen, 1978).
What Should People Be Told Before
They Attend an Assessment Center?
They should be told what the assessment center is
evaluating, the kinds of exercises they will be
going through and the use of the information
obtained in the assessment center. As clear a pic-
ture as possible should be painled for the candi-
dates. They should be made as comfortable as
possible so that their behavior is both natural and
consistentwith what Is being demanded of them.
Are Assessment Centers Related to
Sensitivity Training?
No. Assessment centers measure work-related
skills and do not in any way attempt to create the
MICRON1.111D B1'
JORM MICROLAB
CEDAR RAPIDS DES MOINES
The more the test
situation reflects
the job demands
and calls for the
exhibition of skills
that will be
required on the job,
the better the
ability of the tests
to predict later job
behavior.
2;L6 3S
J
J
r
ASSESSMENT
CENTERS
kinds of situations or the way of interpreting those
situations which would be true of most types of
sensitivity training programs.
What Then Is an Assessment Center?
An assessment center, for whatever purpose it is
to be used, is the application of a specific
methodology to the development of information
about a person's job-related strengths and weak-
nesses. The methodology considers what the per-
son must do to succeed in the real world (Job
analysis), the development of instruments to
measure in a job related way the critical skills
needed to succeed and the training of a panel of
individuals to process the information in a fair and
impartial manner.
An assessment center is a sample of job be-
havior and, as such, is only as good as the sample
resembles the job, the length of time an individual
is observed and the ability of the processors of the
information to be fair and accurate.
What Then Is An Assessment
Center Not?
An assessment center is not personality and intel-
ligence tests. It is not a personal interview con-
ducted by relatively untrained managers. It is not
two leaderless group discussion exercises, an
in -basket and a business game. An assessment
center is not pre-packaged exercises, unless you
can demonstrate the relationship to the job
analysis.
Assessment center technology describes the
system by which information. gathered under
controlled conditions - in which people are re -
1 cluired to perform job related tasks - is used to
select, develop, counsel, career plan and any
other activity which would enhance the fit of the
person and the position and the overall effec-
tiveness of the organization.
References
Barrett. C.J. & O'Leary. B. S. "A ddfeientail predanon
model to moderate the effects of heterogeneous groups in
personnel selection and chssilicanmi', Personnel
Psychology, 1969.22:118
Bray, 0. W & Campbell. R.J. "Selection of salesmen by
means of an assessment center". Journal of Appried
Psychology, 1968,52.36.41.
Bray, D. W & Grant. D, L. "The assessment center in the
meawrementof potential for business management'.
Psychological Monogmphs, 1966,80.117, Whole No. 6251.
Cohen, S.L. '5nndardization of assessment center
technology: some cndeal concerns', Journal of Assessment
Center Technology, 1978, 1f31. 1.10.
Equal Employment Opportunity Commission, Chit I service
Commission, Department of labor. Department of Justice.
Uniform guidelines on employee selection procedures.
Federal Register. Vol. 43. No. 166. Auguu 25, 1978.
Frank. F. & Drucker. J. "The influence of evaluatee's see and
evaluation of it response on a managerial selection
msmumenl".sex Roles: A Joumal of Research 3. 1977.
Griggs vs. Duke PowerCmrgmny, 3 FEP 175119711.
Hicks vs. Crnwrr Zellerbach,'L FEP 1059119701.
Hindchs,.1. R. -An eight year Olnwup of a management
assessment center". Journal of Applied Psychologv, 1978.
63,151596601
Huck, J. R. "A;sessmenl centersa review of the external and
internal validities", Persc rmelPsychology. 1973.26,
191.212.
Huck, J.R. & Bny, D.W. "Managernenl assessment center
evaluatinns and sub..gmmt Inb performance of white and
black Iemsles". Personnel Psy010Irn3y. 1976, 29. 1330.
.Iallee, C. L.. Bender. J. & Calwn. O L. "the usses.ment
center technique: a validation smdv'. Management of
Persannd Quanedv. 1970.90.9-14
Kraut. A 1. Scon. G.J. " Valldlty of an operational
manegemenl assessment prMJntn" Joymnlnf Applied
Psychology. 1972. Vol. 56, No. 2. 124.129.
Lopez. F. M. "Current problems In test performance ul loh
applicant -, Personnel Psychology. 196b. 19. 10 19
binws, J.I. 'Tho devvinpmant of an assessment center for
the idennheVion of agwmarn! potvnlwl".Persannd
Psyncc1Nv. 1973.26.56959I1
Moses..H.&Btcehm.VE."Relationshipofaswssmenl
renter pmtotmance In management progress of.umen '.
Jnnmalnf Applied P.rychology. 1975.60 527529.
hlyan vs Motorola, HOCong. Record 566264119641
Schmie. N, & Hill. T E. "Sex and race composition of
assessment center groups as a determinant of peer and
assessor ratings". Jounml ofApphed Psychology. 1977,
62.261-264
1Vmbois. G.M. "Validation of externally developed
assessment procedures for identification of supervisory
potential". Personnel Psychology. 1975. 28. 77-91
Reprinted with permission from the February issue of Personnel
Administrator 4. 1980, The American Society for Personnel Administration.
30 Park Drive. Berea, OH 44017,
111CROEIU4ED BY
JORM MICR6LAB
CEDAR RAPIDS DES 10L•IES
9635
J
N