CANDIDATE TOPICS FOR ANALYSIS IN THE PRODUCTION SPHERE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP83M00171R002100090012-3
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 12
• Document Creation Date: December 12, 2016
• Document Release Date: October 11, 2001
• Sequence Number: 12
• Publication Date: March 31, 1980
• Content Type: NOTES

.. ................ . ..: ....... ...... ...................
...........: -:...............
-------- -- ---------
Approved ForF elease 2001/12/04: CIA-RDP83M00171R002100090012-3
ftw
31 March 1980
25X1 A NOTE FOR:
From:
Subject: Candidate Topics for Analysis in the Production Sphere
25X1A
1. Coming up with a list of potentially worthwhile
production-related analyses isn't much of a problem. There are lots
of candidates, or so it seems to me. Coming up with an analytical
plan of procedure is, however, an altogether different matter, since
that requires some sense of which topics are most promising (i e.
where to start), and that in turn implies some kind of firm prior
sense of structure. In other words, we need more than a list of
study topics; we need a systematic arrangement, so that we can relate
the individual, necessarily bite-sized, studies we and others do to
the total process, and to each other.
2. This, no doubt, sounds utopian, and I can't prove it isn't.
But to show you what I mean, I've tried to arrange candidate topics
(gleaned from the writings of the DCI, RMS management, the
Congressional committees, and the program managers and monitors)
against a hurried, admittedly crude, characterization of the
intelligence "production" (in an economic sense) process.
Additionally, I've tried to describe my perception of the thrust of
the analysis that might be done on each topic by using three labels:
prescriptive, descriptive and definitional. Prescriptive means the
analysis would be aimed directly at developing and reso ving topical
issues in immediate support of the budget preparation process. B
descriptive, I have in mind more general, less normative, y
statistically-oriented studies designed to provide management with
basic parameters about production activities: what goes in, what
comes out, and from where. These kinds of studies would be more
useful to resource management in a strategic, rather than a tactical,
sense. The short paper I recently did on periodicals is an example
of a descriptive study. Another is'the "External Contract Support"
piece nd I did last year. Definitional studies would be
one st removed from budget preparations. They would aim at
12
------------
--------------
--------------
Approved For Release 2001/12/04: CIA-RDP83MOOl71 R002100090012-3
 Approved P Release 2001/12/04: CIA-RDP83M00U71R002100090012-3
developing the kind of basic understanding of relationships (i.e.,
the structure) whose absent I lamented above. Plain logic would be a
major input in seeking principles, describing process, and defining
terms. My embryonic model of the sequential (conversion, collation,
correlation and projection), analytical phases in the production
process is an example of what I mean by definitional analysis.
3. The candidates which follow, each accompanied by a thumbnail
sketch, are by no means exhaustive, nor is it clear that PAAO is the
most appropriate organization to tackle many of the topics. Some are
probably more appropriate to NFAC or external contractors. My
,feeling is that, whoever undertakes the analysis, the-scope of any
inquiry should be Community-wide.
4. The Process Diagram I've used to try to give some order to
the topics is super-simple, but I think useful as a starting point.
Refinements would make the functions more explicit (and a lot more
numerous), recognize parallel process (i.e., for National and
Tactical, the several collection disciplines and various subjects or
projects within), and accommodate the time dimension.
Approved For Release 2001/12/04: CIA-RDP83MOO171R002100090012-3
 I
I ___- ~ - s ~3vt~f
f
cwre p ~ `/?
~1E INrE ~EWCE p0.0LvS S
A proved.F 004442084. -RDP83M001.Za R002100090012-3
caLVECTOtJ
GAPAc%Ty
T
LoL~ECrEfl
.~NFO.a
MCrMT.
CVNLTIOW
.............. .....
.. ..........
------------------
12
------------
12,
FEE
PA.OtE55 - --_- - -
- ------------------
..::.......:........_..._......
_ .......................... _..........._......................................................._....................._.. "... .............................-.._........ :.. _
R#C ti-t
An4A,.YSTS) tcoWvt J
SUPPORT
A WP..'
ANALYT CA.
fLAI~AAVTY
 Approved For R216ase 2001/12/04: CIA-RDP83M00171F W2100090012-3
Candidate Topics*
25X1A
Function 1: Analyst Recruitment
? Characteristics and Qualifications of Production Analysts
Assemble community-wide inventory of analyst
professional traits. Contrast current analyst
population with past populations; other government
professionals; professionals in kindred research
fields. (Descriptive)
uduu
? Community's Competitive Position re New ads
Analyze Community's ability to attract highly
qualified graduates. Develop and cost proposals for
enhancing recruitment practices if applicable.
(Prescriptive)
Function 2: Analyst Training and Support
? Programs to Enhance Analyst Training and Support
Draw a profile of the non-ADP programs, planned and
underway, to promote analyst skills, expertise, morale
and motivation. (Descriptive)
NOTE: This study would pursue the objectives sketched
in memo of 13 November 1979.
? Value of Alternative Support Measures
Present programs include a wide range (sabbaticals,
rotational assignments, incentive awards, etc.) of
intuitively plausible techniques for promoting analyst
performance. What is missing is any set of standards
for gauging how much should be invested in each. In
the absence of this kind of understanding, a little
bit of everything gets tried. Somewhere out there,
there must be a body of literature (hopefully
including empirical studies) that addresses the
*Keyed to attached process diagram.
Approved For Release 2001/12/04: CIA-RDP83MOOl71 R002100090012-3
 Approved For Release 2001/12/04: CIA-RDP83M00171R002100090012-3
relationship between such measures, including the
physical working environment, and average levels of
professional performance. The suggestion here is to
search for that information, identify it, and
summarize the findings applicable to supporting
intelligence analysts. (Prescriptive)
NOTE: This is probably a contract item.
Function 3: Use of External Contractors
? Level of Usage
Document the Community's current use of external
contract support; contrast with past levels. Identify
motives for using contractors (no in-house capability,
desire for independent perspective, etc.).
(Descriptive)
NOTE: PAO did something like this last year to
respond to SSCI interest.
? Guidelines for Employing External Support
Formulate general principles governing the choice
between internal and external analysis.
(Definitional)
NOTE: This may be only a subset of the Resource
Allocation Principles topic suggested below under
Function 4. The issue of competing analysis is
similarly a related one.
Quantity A:
Analtyical Capacity
? Measurement Methodology
Is there any reasonable way to quantify and compare
the Community's capacity to produce intelligence on
various subjects, in various geographic areas of
interest. Existing management information systems can
provide approximate data on the dollars and manpower
that is being put into a particular subject-region
------- ------
................
--------- - --
 Approved FoRelease 2001/12/04: CIA-RDP83M0017JR002100090012-3
sector of the production organization, but can we
implicitly assume that this relates in some homogenous
way to our capacity to make things come out of that
sector? Are some sectors inherently less efficient at
transforming resource inputs to production capacity
than others? If so, this should be recognized and
accommodated in the resource allocation decision
process. (Definitional)
Function 4: Production Management
? Topic Selection Mechanisms and Procedures
Document geneses of various analytical products.
Discuss pros and cons of self-initiated studies.
Describe Community-mechanisms for co-ordinating
production activities (necessary to assure that
parallel analytical efforts proceed only as a result
of deliberate choice). (Descriptive)
Producer - Collector Interactions
Identify major lines of communication between
production sectors and their principal sources of
information in the collection sphere. Survey
producers for insights concerning bottlenecks to the
timely flow of information and how they might be
alleviated. (Descriptive - Prescriptive)
NOTE: Pilot survey now being conducted by
may contribute toward appraising feasibility o
relating production activities to their collection
sources.
? Resource Allocation Methodology
Endeavor to set down a rigorous, truly cogent
rationale for telling good redundant analysis from bad
redundant analysis. Convert this conceptual
foundation into plain English rules-of-thumb for
guiding production managers in their decisions
concerning commitment of the scarce analytical
resources they control. (Definitional)
NOTE: This would expand on the analysis PAO did in
response to the HPSCI request for such a rationale.
------------
Approved For Release 2001/12/04: CIA-RDP83MOOl71 R002100090012-3 "----
-- ----------
_ ............. _......................... .................................... ................. . --
...........:........ .....,..........~:..........................- ........................................................-- ....
--------- ------
25X1:A _
 Approved F& Release 2001/12/04: CIA-RDP83M00:j71 R002100090012-3
Quantity B: The Supply of Intelligence
? National Intelligence Product Line
Survey national intelligence producers for up-to-date
data on their output: names and types of
publications, distribution lists, release frequency,
and the like. This study would be patterned on the
PAID projects of previous years, with certain
methodological improvements and, optimally, expanded
to include the full range of trans-departmental
products, not simply periodicals. (Descriptive)
NOTE: While of general interest value in its own
right, this study would also be useful as a
springboard for more ambitious efforts to relate costs
to products, or to evaluate consumer satisfaction, if
such studies were deemed worthwhile.
Function 5: Marketing and Product Evaluation
? Supply-Demand Correlations
Document and compare ten-year trends in the subject-
regional demand for intelligence products, as
reflected in changing patterns of DCID 1/2 priorities,
and the allocation of production resources, as
recorded in CIRIS. (Descriptive)
NOTE: I plan to write up this short study at the
first opportunity. Most of the analysis has already
been completed. A major caveat is the implicit
assumption that what goes in (i.e., CIRIS dollars) is
proportionate to what comes out (i.e., volume of
intelligence products). See comments under "Quantity
A"
? Practical Applications of Formal Priority Statements
Evaluate compatibility of subject emphases in the
different (DCID 1/2, NITS, KIR) priority statements
(i.e., demand proxies) now in existence. Survey
procedures to ascertain impact of these documents in
resource commitment decisions. Seek suggestions
Approved For Release 2001/12/04: CIA-RDP83M00l71 R002100090012-3 - =
. - - ----------------. - ................ ............. -....-----------------------.-_..-......_...._............
................. --...._.._.._.....~ ._..........
............................................................................................................................................................. ................. ...................... ................... .....................................
 ---- --- ------ ---- -
Approved FoTRelease 2001/12/04: CIA-RDP83M0017JR002100090012-3
for improving their applicability to this purpose. If
there appear to be inherent major limitations with
formal priority statements, what alternative methods
might be employed to effectively communicate consumer
interests to intelligence producers? Explore the
suitability of incorporating expressions of the demand
for intelligence in the program guidance, and in the
budget deliberations. (Descriptive - Definitional)
25X1A
? Mechanisms and Procedures for Product Evaluation
Ascertain how the Community now goes about evaluating
the quality of its product, and the satisfaction of
various consumers with it. Evaluate the desirability
and feasibility of establishing mechanisms for
effecting product evaluation and providing consumer
feedback on a sustained basis. What standards would
be applied? Are there intrinsic qualities to an
analytical product, separate and distinct from the
pleasure with which it is greeted by the consumer,
which make it good or bad? (Definitional)
NOTE: The urvey previously cited may also
contribute to un ers anding what, if any, differences
might exist between analysis that is "good", and
analysis that is "successful".
Function 6: Distribution of Collected Information
Influence of Impediments to the Distribution of Collected
Information on the Quality of Analysis
Congressional committees have raised the
compartmentation issue often. Additionally, there
seems to be some sentiment among analysts that
agencies tend to favor their own production components
with preferential access to the information their
collection components have acquired. How real is the
perennial compartmentation issue (i.e., do the
analysts themselves feel it is an impediment? Are
informal channels available that mitigate its
influence in denying information?), Are the
competitive institutional obstacles important? What,
if any, changes might facilitate the productive flow
of inter-producer information? (Descriptive -
Definitional)
Approved For Release 2001/12/04: CIA-RDP83MOOl71 R002100090012-3
12
 Approved For Release 2001/12/04: CIA-RDP83M00171R002100090012-3
Process-Wide
? Identification of Resource Bottlenecks (via case studies)
This effort, or series of efforts, would examine
recent instances of particularly acute criticism of
Community performance in search of examples of where
the availablity of more or different resources might
have had a significantly beneficial effect. Delays in
detecting a Soviet combat presence in Cuba, and North
Korea's build-up of its ground forces, are two prime
candidates for such analyses, which need not start
from scratch (as in the traditional sense of a "post
mortem"), but would rather draw heavily upon already
completed investigations and reviews. The emphasis
would be on specific resource solutions to what went
wrong, and the general applicability of those
solutions elsewhere in the production Community.
(Prescriptive)
? Economic Model of the Intelligence Production Process
Studies such as the search for specific resource
bottlenecks, above, presume some prior understanding
of the sequential functions which constitute the
production (in the broad, economic sense) process.
While the existence of such a model is implicit in
this and other topical issues, such as the balance
between collection, (data) processing, and production
(in the less broad, intelligence usage), the
understanding has never been made explicit. Achieving
a rigorous comprehensive representation of the
intelligence process would be no mean undertaking, but
basic research along this lines could, if successful,
provide a fundamentally useful and enduring tool to
successive generations of Community and program
managers, not to mention production/analysis analysts.
(Definitional)
NOTE: The model used to organize this paper is a
rudimentary example of the kind of thing this study
would seek to achieve.
.........
APpraovod..FA.r.Releas~e_2001/12/.0.4..:..C