ASSESSMENT CENTERS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79-00498A000200120020-4
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 9
• Document Creation Date: December 12, 2016
• Document Release Date: February 13, 2002
• Sequence Number: 20
• Publication Date: December 1, 1976
• Content Type: MF

IDD/A _...:nistry
Approved For Release 2002/03/25 : CIA-RDP79-00498A20012002014
MEMORANDUM FOR: Deputy Director for Intelligence
VIA
Deputy Director for Administration
John F. Blake
Charles A. Bohrer, M. D.
Director of Medical Services
SUBJECT
: Assessment Centers
1. In the continuing search for ways of improving
Agency management, the potential of assessment centers
for identification and development of managerial talent
receives periodic reexamination. At the request of the
DD/A. I recently reviewed with him our activities in this
area within the Directorate. In the course of our dis-
cussion, he observed that other Directorates may want to
reconsider at this time the possible application of this
method and suggested that I bring the concept to your
attention as one worth looking into.
2. The Management Committee discussed the subject
about a year ago but took no formal action. A background
paper prepared prior to that discussion is attached.
Should you have any interest in exploring this further, Jack
Blake and I and members of my staff would be pleased to
respond.
CHARLES A. EoiiiiaER, M.D.
Charles A. Bohrer, M. D.
Attachment
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
The Assessment Center Approach to the Identification and
Development of Managers in CIA: A Background raper
OCT T915
Typically in the public service, the process of identi-
fying employees as possessing managerial potential is based
largely on their past performance in a technical specialty.
One strategy for overcoming the inherent.limi.tations of this
approach involves the use of multiple assessment techniques,
especially those involving a. simulation of conditions and
problems which the manager actually confronts on the job.
Properly designed, such techniques can permit an evaluation
of managerial skills by replicating those interpersonal,
administrative, and decision-making demands unique to that
particular organizational environment: in which the manager
must function. Integrated with traditional. information inputs,
including supervisors' evaluations and the performance record,
such data can materially enhance the predictive accuracy of
managerial selection, identify individual development needs,
and enlighten senior management on the characteristics of its
managerial pool. in relation to projected requirements and
future demands. In sum, such techniques offer management an
additional tool for predicting and directing the capabilities,
character, and style of the organization.
In August, 1972, a journal article setting forth these
points came to the attention of Mr. Colby, then Executive
Director. Forwarding the article to the Deputy Director for
Support, Mr. Colby made note of the fact that OMS psychological
assessment is focussed primarily on initial selection and early
career guidance, and added that the article raised in his mind
"the possibility of a 'mid?-career' aassessmen.t to identify strong
and weak points at that stage, for training and leadership
development purposes if not for raw selection".
At this point, the OMS Committee for the Behavioral Sciences
had already devoted a series of meetings to this same general
subject and had. reached a similar conclusion. In responding to
a request for comment on Mr. Colby's note, the Director of
Medical Services concluded in a memo to the DDS that "ewe should
try the Assessment Center technique in the Agency".
Subsequently, the DDS announced that OMS was prepared to
discuss the concept with interested offices, and to render
assistance to any office which chose to develop a Center. The
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
Psychological Services Staff was designated as the resource
for such development, and this responsibility was later formal-
ized as a Directorate-level objective within the MBO framework.
Discussions with several offices ensued, and eventually a
developmental effort was begun in the Office of Joint Computer
Support. An OJCS Center was established and continues, and a
similar effort is presently underway in the Office of Communi-
cations.
As with any new venture, many problems arose between the
idea and its fruition. Progress has been painfully slow. Ex-
perience to date affirms that the concept is basically sound.
Full understanding of the concept and anticipation of some of
the problems at the outset are essential in evaluating its
potential for any given application. The balance of this paper
attempts to deal with these matters in some detail.
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
1. A brief him While the pioneering efforts of the
OSS Assessment Staff had been preceded by earlier work in
Britain (the War Office Selection Boards) and. still earlier in
Germany, it was the OSS experience which most profoundly in-
fluenced the later development of assessment methods for organi-
zational applications. Concerned solely with selection, the
OSS effort was the first large-scale effort to incorporate the
use of multiple measures and observations and multiple assessors
in a single assessment exercise, relying heavily on unique situa.
tional tasks which simulated job conditions, and aimed at a
global judgment of the candidate's assets and liabilities-for
the job in question.
CIA continued this tradition on a reduced scale. As
the volume of. candidates declined after the early 1950's, assess-
ment gradually shifted from a group process to a one-assessor-
one-candidate interaction, with less use of simulation exercises
and heavier reliance on the standard psychological measuring in-
struments appropriate to the one-on-one assessment condition.
Exceptions continued in the case of programs where highly special-
ized selection was required (e.g., pilots for the "I" and "0"
programs). Here the existence of a pool of pre--selected candidates
with unique qualifications for demanding tasks permitted the use
of group techniques and required the use of devices carefully-
hand-tailored to make the difficult discriminations required
between individuals who were all highly able and possessed common
backgrounds of outstandingly successful. achievement in their
special field. In such cases, assessment adhered more closely
to the OSS model. In general, however, assessment today by the
Psychological Services Staff most typically consists of a candidate
being seen for one day by a psychologist who, often having already
in hand the results of the candidate's performance on the Pro-
fessional Test Battery, uses this ti.mc~ to obtain additional data
from interviews and written procedures to put together a picture
of the psychological factors which govern the candidate's behavior.
With this picture in mind, he then prepares a, narrative report,
descriptive in nature, geared to the suitability questions and'
concerns of the referring office.
In addition to the selection question which governed
the OSS operation, users of Agency assessment services today are
often concerned with questions of assignment, career direction,
and potential growth as well. It is still true, however, that
assessment is a management device used most often in CIA for
applicants and for employees in the early stages of their careers.
The relatively few persons seen at later career stages are most
often referred because "problems" have arisen and assessment is
seen as a technique to help figure out what went wrong and what
to do about it.
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
In other quarters, meanwhile, the OSS model triggered
a rather different set of developments. Following some unsuc-
cessful post-war attempts to apply OSS-type assessment methods
to selection of candidates for specialized professional training
programs, the approach fell into temporary disfavor. Ten years
after OSS, one of the most prominent American psychologists
declared that assessment in the OSS tradition was a proven
failure. While this death notice was remarkably premature,
there were few to dispute this judgment at the time. Few if
any major assessment programs were in operation during the 1950's
outside of .the CIA.
However, it was in 1956 that American Telephone and
Telegraph quietly embarked on a program which was later to prove
how wrong this judgment was. AT&T launched a long-term, ambitious,
carefully controlled research effort to study the management de-
velopment process in the Bell System and to identify the variables
related to managerial success. Assessment methods were incorpo-
rated into the basic research strategy. It was several years
later before the results came in. When they did, the data
demons_trated.conclusively.that OSS-typo assessment methods--could
identify managerial potential, and do so more effectively than
conventional methods relying exclusively on job performance in
sub-managerial positions. This finding led to adoption of an
assessment approach (now dubbed "the Assessment Center") through-
out the Bell System.
When news of this success spread, similar efforts were
soon launched in IBM, General Electric, J. C. Penny, Standard Oil,
and other prominent companies eager to emulate the AT&T success.
Literally hundreds of other organizations followed suit, including
several government agencies, among the first of which were Internal.
Revenue, the U. S. Forest Service, TVA, and FAA. Not all of these
applications were marked by the kind of careful research effort
needed to establish their validity and usefulness. But many of
them did, and by now the volume of research evidence testifying
to the soundness of the method in a wide variety of settings is
sufficient to suggest that almost any organization intent on
improving managerial selection and development should give the
concept serious consideration.
2. Description of a t_cal Center. An assessment "center",
of course, is not a place but a process. The process involves
the systematic application of a set of procedures for evaluating
people. The procedures are devised and selected to have the
maximum possible relevance to the managerial job on which the
assessment process is focused. The process consists primarily
of placing individuals in problem situations--simulations of
real-life tasks--which they must handle, while their behavior
is observed and evaluated by a team of assessors. The situational
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
tasks are developed jointly by psychologists and managers to
reflect key aspects of performance in the managerial job in
question. Typically they include an in-basket exercise, a
leaderless group discussion, and one or more management "games",
all designed to elicit managerial thinking, decision-making and.
action. Assessors are drawn from line management of the organi-
zation, typically two organizational levels above the assessees,
and trained from two to five days in how to observe and judge
behavior in the various assessment exercises. Assessment is
conducted off-site and away from the work environment, over a
two or three day period. Generally, no more than six people are
assessed at a time, with one.assessor for--every one or two assess-
ees. In-depth background interviews, peer evaluations, and video
recordings and playback are also include(] in the process. A
profile of strengths and deficiencies on each of the dimensions
previously identified as important to job success is generated
for each participant. Performance is not judged as pass or fail;
it is rated rather than ranked, and feedback to the assessee is
provided orally or in writing. A written evaluation is made
available to the appropriate persons in the managerial structure,
varying with the basic purpose of the assessment center, which
may focus in varying degrees on selection, training needs, or
other aspects of career management and development. When develop-
ment rather than selection is the primary goal, the process is
sometimes referred to as the "Management Development Center"
rather than "Assessment Center"--a perhaps more acceptable as well
as more accurate term. The OJCS Center has been so labelled.
3. Steps and problems in implementation. While the de-
scription above outlines an essentially simple mechanism, some
reflection on the implications will surface the fact that the
assessment center is actually a profound departure from tradi-
tional approaches to assessment in both the OSS and the CIA. It
differs not only in focusing on managerial selection and develop-
ment at a later career stage, but in more fundamental ways as well.
The OSS Assessment Staff was a, semi-autonomous unit,
staffed by experts in human behavior, which developed its own
standards and delivered candidates into the system judged by
those standards. The Assessment Branch of the Psychological
Services.Staff is likewise a group of experts who, operating in
a consulting capacity, feed into the system interpreted data and
their professional opinions and judgments about individual assets
and liabilities. They do this by relating their psychological
conceptions of the individual and their knowledge of predictive
relationships between psychological data and everyday behavior
to the general domain of job demands as they understand them.
The assessment center, in contrast, proceeds from a quite differ-
ent set of assumptions.
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
First in importance is the assumption that the require-
ments of a given managerial position, for example, that of Branch
Chief in a given component, constitute a set of demands unique
to that position which can be (a) identified by careful job
analysis and (b) once identified, targeted by specific assessment
devices which will elicit behavior relevant to those demands.
It follows from this that the first order of business in planning
an assessment center is for the office concerned to decide what
target positions should be the focus of the effort, after which
a detailed process of analyzing what the incumbents of those
positions actually must, do to. perform .effectively.b.egins
In some cases, this step is slighted, based on fairly
valid reasoning which holds that all managerial positions embrace
certain common elements: oral and written communication skills,
energy, flexibility, creativity, etc. In our experience to date,
short-circuiting of this step'is ill-advised. While such common
elements surely exist, their relative importance varies consider-
ably. For example, careful job analysis in one office established
the fact that "ability to delegate" and "keeping others informed"
were critical job elements at the top of the list--far more im-
portant than, for example, flexibility and tolerance of stress.
For managers at the same. level in another office, this order
might well be reversed.
A second assumption of major importance is that once
a center has been properly designed to elicit job-relevant be-
havior from the assessees, line managers, not psychologists are
the appropriate persons to judge the effectiveness of that be-
havior. (This assumption is supported by research indicating
that trained managers do at least as well as psychologists in
predicting later managerial performance from assessment center
data.) Using line managers as the assessors takes the psychol-
ogists out of one of their traditional roles but continues to
use them where their skills are highly relevant: in job analysis,
construction of assessment devices, training assessors, con-
solidation of the data, and as consultants in the overall operation
of the center. In their training fuction, the goal is specif-
ically not. to make psychologists out of. the assessors, but rather
to sharpen their objectivity and systematic approach to observa-.
Lion, recording, and integration of the behavioral data,
From this shifting of roles comes also one of the major
fringe benefits of the center, namely, that which accrues to the
manager from his experience in serving as an assessor. OJCS
assessors extol the value of this experience, and General Electric
feels so strongly about it that it has established'a policy of a
1-to-l assessor-candidate ratio and rotates managers in the
assessor's role to expose a substantial percentage of its managers
to this experience.
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
A third major assumption implicit in the process de-
scribed is the necessity of incorporating the assessment center
as an integral part of the personnel management machinery. Most
Agency components already have a variety of mechanisms--career
boards, promotion panels, annual evaluation procedures, etc.--
some quite elaborate and functioning very well. To embark on
an assessment center operation without carefully thinking through
where it is going to fit into the system and how the output of
the center will be integrated with the function of these other
mechanisms is, at best, an invitation to resistance, confusion,
and inefficiency, and at worst, a blueprint for disaster.
While this may seem obvious, one of the factors which
makes it less so is the long tradition in CIA of regarding assess-
ment as something that is done by "the experts"--an outside group
to whom you send people and get back some hopefully useful in-
formation about them. At first blush, it may appear that the
assessment center also is something that "the experts'" are going
to provide, with the added disadvantage of requiring a heavier
investment of time and effort by the user office than before.
The idea that the center, if it is to come about, must be "built"
by the office itself, and once built, "belongs" to the office in
the most literal sense, with the "experts" only advising and
helping in its creation and functioning, is not all that readily
perceived at the beginning. Coming to grips with these issues at
the outset is essential to realistic appraisal of the pros and
cons of embarking on such an effort, and greatly enhances the
chances of a successful outcome if one decides to proceed.
4. Considerations for the future. The preceding discussion
has attempted to outline some of the problems and difficulties as
well as the benefits of using assessment centers. They're ex-
pensive. Resources to mount such efforts are limited. In the
more than throe years since the idea was first seriously broached
in the Agency, progress has been slow. But they work. And they
work here. Improved managerial selection and development can
result at almost any managerial level.
In 1973-74, assessment centers were used for the first
time on a government-wide basis in the final selection of 25
candidates for an executive development program open to nearly
all CS-15's in the government. The Office of Management and
Budget, in cooperation with the Civil Service Commission,
sponsored this program, which eventually resulted in 100 selected
applicants participating in an assessment center conducted by 50
high-level government executives trained in an 8-day assessor
training program. The assessment center included a government
in-basket, a group discussion simulating the cabinet of a hypo-
thetical country, a group discussion on national priorities, a
mock press conference, and an analysis problem involving the
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
 Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4
staffing of a government agency. 0MB expressed the hope that
this effort would be a prototype of a much larger government
.effort in this area.
Use of assessment centers with minority personnel
and women have had favorable results. In one landmark decision
involving litigation between AT&T and the EEOC, a consent decree
provided that ATF,T. open its assessment centers to 2,000 female
college graduates in lower level. management positions.to de-
termine their potential for higher level management positions.
'Other experience in the Bell System demonstrates that' their -
assessment centers operate fairly for blacks as well as whites.
A recent communication from 0MB to the Comptroller called
upon the Agency to see that budget estimates "reflect full con-
sideration. of---the resources needed for Agency systems to identi-
fy, develop, and utilize career executives." Assessment centers
would clearly appear to qualify as one such resource.
In addition to internal Agency resources for hastening
the development of assessment centers, a number of external re-
sources are readily available through consulting firms who special-
ize in this activity. Should the Agency decide to implement this
approach at a faster rate than could be supported solely by in-
house resources, the Psychological Services Staff is prepared. to
identify and recommend such outside source"3 of-help.
Approved For Release 2002/03/25 : CIA-RDP79-00498A000200120020-4