UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, C.C. 2046)
September 27, 1979
Honorable Douglas M. Costle
Administrator
U.S. Environmental Protection Agency-
Washington, D.C* 20460
Dear Mr, Costle;
On May 30, 1979, you met with the Environmental
Measurements Committee of the Science Advisory Board. At that
tine and in a subsequent conversation with Dr. Douglas Seba, the
Executive Secretary for the Committee, you expressed a keen
interest in and a personal frustration with the problems of
information management within the Agency. We have looked at
various problems and reviewed outside criticisms (Appendix A)
and, as a result, offer you our suggestions, which are outlined
in the attached report. We are interested in meeting with you
or your designees to elaborate on our suggestions and to discuss
alternatives, should you so desire. We recognize that both the
good reputation and success, as well as the morale, of the
Agency rest heavily on the types of decisions to be made in
carrying out constructive and corrective actions for information
management.
Our primary recommendation is that the Agency develop an
overall, Agency-wide information policy. From the•information
thus far received the Committee concluded that this overall
policy must be endorsed and given managerial authority by the
Administrator of the Agency* Further, the Committee feels that
encouraging cooperating divisions to mutually consider and
develop an overall policy has not worked in the past, as
indicated by plans and suggestions offered to the Committee by
documents and by Agency staff and consultants (Appendxes A and
B). Consequently, the Committee has suggested alternative ways
of developing and implementing an overall policy, as are also
outlined in the attached report.
Basically, the data collected by the Agency is program
(i.e., source}- oriented, toxics data in one section,
permits data in another, and. water quality data in still another
section. The Committee feels strongly that this system causes a
serious lack of logic and coordination of EPA's data. To
overcome this problem, the Committee suggests that under one
policy, the data be arranged not according to where it came
from, but according to how it relates to pressing environmental
problems. Such a system, which is the result of overall
guidance and policy, would be problem-oriented rather than
program- oriented.
-------
Employing this as a starting point would not only best
serve all types of users in the country, but would best serve
the Agency as well. It is. the type o£ nucleus upon which the
most efficient and cost saving system can be constructed, since
its objective is to answer the questions most often posed by
critics. The Committee requests that the Administrator take our
recommendations into consideration.
We wish to emphasize that the avoidance of parochialism in
developing and implementing such a policy is of the greatest
importance. You may note the failure of information
coordination among cooperating divisions in the past, as well as
the inaccessibility of much of SPA's data. We do not recommend
the use of existing administrative councils for the same reason.
Our suggested alternatives for policy execution underscore the
need for an individual manager responsive to the needs of
.users and devoted to the purpose of overall information
management in the long run, as well as the short term.
The overview is respectfully submitted to you accompanied
by our strong belief that a progressive solution to the issue
can be attained within the Agency. We1re looking forward to
your reply.
Sine
Leonard Greenoideld ,' Chairiru
Environmental Measurements Committee
Science Advisory Board
Attachments
-------
Environmental Measurements Committee Overview and
Recommendations for Solving Information
Problems EPA.
All decision-making requires basic information which is
pertinent and of high quality. We deliberately emphasize the
phrase "all decision-making" and then give weight to the fact
that responsibility for the correct application of data lies/ or
should lie, within every administrative, congressional, and
judicial office as well as in offices in the private sector. If
decision and action are not based on information, high-
probability-of-chaos will be the important statistic of the
future. Indeed/ some of the existing problems in government/
the results of which are directly visible to or consequent upon
a sensitive public, roust be due, at least partly, 'to a famine of
pertinent information,
This famine of information may be caused only in part by
insufficient information. In fact, an equal or greater problem'
may be that of finding and applying the data available. Imagine
the trouble, expense, and thought going into the collection of
facts needed to make an important decision, only to have it lost
because of problems in data storage and retrieval.
Despite the requirement of information for input into
decision-making, the government, for the most part, does not
assign any priority to the need for information — much less a
high or low priority. It does not assign it a budgetary
importance; many instances may be given of agency budget cuts
borne most heavily by information branches. It pays no heed to
the extreme need for the organization, centralization, and
distribution of data within its purview* The government's
attitude is cavalier, at best, towards the'requirement of
specific information sought by citizens acting as private
users.
In EPA, this problem is very evident, and both commission
and ommission errors have 'been elaborated on in a number of •
documents (Appendix A), along with curative suggestions and
suggested mechanisms* These documents orginate both within the
Agency and from outside communicators and consultants
(Appendix B). (Some of the consultants are former EPA employees.)
-------
Notw,1' ^standing such documentation, the first step in
resolving _ne problems associated with the production and
management of information in the Agency must be an
acknowledgement and recognition that the problems exist. Outside
critics have been insisting so for several years, but the nature
of these problems have not been well understood or clearing
defined.
Second, the Agency must agree upon the nature of the
information needed. Instead, the definition of what is needed,
in the past, has been left initially to the producers of data or
to ADP consultants setting up data base systems. This
internalized thinking has failed fundamentally to recognize that
information encompasses a great deal more than what exists in
formal information systems. The documentation referred to in
Appendix -does not indicate that any formal canvass has ever
been made of users of data. Users are the public. They consist
of office and laboratory personnel within the Agency, Federal,
state, local, and regional government officials and their
technical staffs, and individuals and groups in the private
sector. Users can also be the producers of information» We have
data from bureaus and offices that relate the numbers of
inquiries successfully answered each year—questions that come
from all kinds of users. Without belaboring the area of
numerology, EPA should be receiving and answering queries in
quantities far beyond the numbers indicated by such EPA data.
EPA should have the reputation for doing so, thus becoming an
Agency sought after by users as one that is both good and
reliable. Because it is not, EPA must recognize that the concept
of information and information management is not understood by
the senior management of the Agency. In our opinion, this is
one of EPA's major deficiencies with regard to this problem.
Users are the final pudges of the success or failure of any
information system design, implementation and evaluation, and
yet SPA has not considered the users as important.
Logic then leads us to inquire about the attributes of a
good information system; What are the types of questions to be
answered? What are the requirements for pertinent data of high
quality? Sow may the data be stored best so that answers to
questions may be easily obtained? How may the user learn the
proper language to obtain the data? All are fundamental
inquiries into the process and success of the system. Yet we
have little evidence that these questions have ever been asked
by senior Agency management.
-------
A slightly different point of view is that an good agency,
with a good repretation, which answers its mission successfully,
is one whose information-distribution system is a natter of
pride. In fact, the success of the agency can be directly
related to this and is often judged by it. In effect, the.
information-distribution system should be a showcase of how data
are to be taken and dispensed. Instead, critics (OMB, OMB,
Congress, NAS) constantly complain about the paucity, quality,
and inaccessibility of EPA*s data.
We believe that the positive road toward solving this
problem is first the. establishment of an Agencywide information
policy. The recent efforts of the DAA Advisory Group on
Monitoring and Information Management is a first step, but only
a first step. We feel that a longer-range solution must
encompass the following recommendations. It is not our purpose
to elaborate on such a policy, only to urge its adoption by the
Administrator* These recommendations and alternatives are
presented for- consideration at the strategic level, based upon
discussions among ourselves and with consultants working with
us.
The strength of the proposed information policy is thus
partially defined by the following;
1. It shall be Agencywide.
2. It will be principally defined by the
ultimate users of the system.
The following recommendations ask some basic questions. The
answers to these questions will reveal what previously has been
overlooked or ignored and will/ in turn, dictate exactly how the
policy must be shaped.
Many of the recommendations are discussed in the appended
documents. (They should be examined first before any of the
following are treated as new recommendations".) The
recommendations in this document vary in that they are based on
the principle that the users of.data will dictate the type of
systea(s) and policy to be devised. Most of the appended
documents, though perhaps stating otherwise, do not deal with
the problems from the viewpoint of the user, as they do not
understand the nature of the user problem.
-------
Before presenting the recommendations, however, a comment
on the term "data" Is called" for. One part of the data recovery
problem is that data are oft.ii treated as isolated facts. In
reality a datum is part of a continuum within some system.
Within the Agency, data are not the private purview of any group
or individual and, therefore, should not be defined by any one
group or individual. Treating or defining data in isolation not
only leads to duplication and frustration for the user, but also
treats the "rights" of the user in a cavalier fashion. Once
defined and collected, the data become the responsibility of the
Agency, which must insure that it is properly stored as part of a
continuum and ultimately distributed to those who need it. That
mandates a basic change in philosophy for the Agency, namely,
again, the recognition that all data, even those seemingly
unrelated, raust be integrated and controlled under a single
policy.
Suggested recommendations for the establishment of a
coordinated information management policy are as follows;
1. Survey past information data plans that have been proposed
to determine their merit. Look at case studies that may
have been presented. Ascertain their feasibility and what they
have in common. Develop a plan for integrating information
management and monitoring data on an Agencywide scale.
2, Second, make a formal canvass of user needs. Ask users,
particularly at the Administrator and senior management
level, what specific current systems they employ to get
their data and have them put a value on them. Ask them
about the type of documentation library they prefer. Survey
both users and producers of data to learn what kinds of
information they need and their patterns of usage in terms
of actual examples. Compare' user^def ine<3 needs with
plans developed in step 1,
3. Determine from users inside and outside EPA what specific
uses an<2 types of questions they adress to EPA, which are
and are not answered by the present information systems.
4, Review reconroendations made by other agencies.
-------
5, Use the revised information management plan to develop an
on-going policy with end-users as its pivotal base. The
plan should include periodic re-surveying of users to
reflect changing needs and priorities.
The Committee suggests that the following be considered as
a trial for establishing a new central information policy and
its mechanisms:
1. Develop policy level support from the Administrator
to insure that there is compliance, cooperation, and
enforcement for the trial of pilot systems or demonstration
projects.
2. The plans for pilot systems should be presented in the
form of seminars to the people, at specific levels/
who are responsible for policy execution, with sufficient
discussion and feedback to be sure they understand the
principles involved,
3, Select an office or Agency branch to try a pilot
operation and test for' success, failure, or alteration
to fit.
4. Select a Coordinator (we strongly suggest one individual
rather than a coordinating group) who will act in the
capacity of policy manager for the pilot and for demonstra-
tion projects. Such an individual should make sure
standards are developed, urge cooperation among program
decisions and act as an ombudsman for the Agency and users.
Ultimately, this coordinator's job should evolve into the
job of an assistant administrator responsible for
information within the Agency. This person would be
responsible for accomplishing all efforts to conform with
the information policy generated by the Administrator.
These efforts -would include; establishment of standards;
coordination of all systems endeavors; development and
maintenance of a cooperative agency policy; control of
appropriate resources such as budget and personnel; and
management of information policy for program divisions such
as libraries, publications, monitoring and data collection,.
ADP and distribution.
-------
The above trial project would involve primarily EPA
personnel and would also help refine and develop the newly
established information policy and its mechanisms, Once a
policy is agreed upon, there must be a proclamation from the
Administrator that the authority lies with a single manager for
enforcing the policy Agencywide. Part of this authority is
within a specific budgetary line. It connotes the ability to
offer funds, in return for which the grantee must conform to
whatever is necessary to keep up the quality and currency of the
data. Unless backing for this type of managerial authority is
obtained from the Administrator, all efforts toward a
coordinated system would be largely wasted.
From the information thus far received (appendix -),
the Committee concluded that there first must be an endorsement
of this managerial authority by the Administrator. Only _t_hen may
efforts proceed toward a coordinated system. Encouraging
cooperating divisions to mutually consider and develop an
overall policy has not worked in the past, as indicated by plans
and suggestions offered to the Committee by documents and by
Agency staff and consultants.
Employing this as a starting point not only best serves all
types of users in the country, but best serves the Agency as
well. It is the type of nucleus upon which the most efficient
and cost saving system can be constructed, since its objective
is to answer the questions most often posed by critics. The
Committee requests that the Administrator take these
recommendations into consideration.
-------
Appendix A
Documents Consulted
ADP stage assessment and five-year ADP management plan.
Briefing book, April 9, 1979. Nolan Norton & Co. One
Military Drive, Lexington, Mass 02173. Also final
presentation revised May 15, 1979.
EPA Blue Ribbon Monitoring Group. February 14, 1979, Technical
support document for specific study. Ensuring that EPA
monitoring data are available to all parts of EPA and ensuring
that area-wide monitoring surveys are tracked in a central
location. Internal document.
Options to improve management of Information Technology with the
Federal Government, Consolidated Information Technology Report,
Executive Office of Administration. Portions A-21-23.
User values in the selection of information services. Final
report on NSF Contract C 1027. Homer J, Hall, Exxon Research
Engineering Co., N.J, 07036, 1977.
Development of management specialisations. Commission on
Federal Paperwork, IRM Report, Sept. 9, 1977,
Needed; A comprehensive information management act. Forest
Woody Horton, Director, Information Management Study, Government
Data Systems.
A study of CSINi The Chemical Substances Information Network
Report.
NiH-EPA Chemical Information Systems, Report #9. June 30, 1979,
Environmental/Chemical Thesaurus, June 1978. Oak Ridge
National Laboratory. ORNL/EIS-132.
U.S. Directory of Environmental Sources (3rd Edition), Jan.
1979, National Focal Point of the United Nations Environment
Program/International Referral System for Sources of
Environmental Information (INFQTERRA),
Guide to EPA Libraries. July 1977. EPA Lib-77-02.
AEROS: Aeronietric and Emissions Reporting System, 'EPA Office
of Air and Waste Management Office of Air Quality Planning and
Standards RTP, N.c. June 1978..
-------
Perspectives on Technical information for Environmental
Protection. Volume I: Analytical Studies for the U.S.
Environmental Protection Agency. National Academy of Sciences;
Washington, D.C. 1977.
Development of Marketing & Recommended Outputs for ORD's
Technical Information Program. Volumes I and II. Final Report
Calculon Corp. Contract No. 68-03-2522 June 2, 1978.
Data Backgroud Paper. EPA Monitoring and Information
Management. Internal Paper Program Evaluation Division, June
21, 1979.
Better Information Management Policies Needed; A study of
Scientific and Technical Bibliographic Services. Report to the
Congress by the Comptroller General, General Accounting Office,
August 6, 1979.
NOAA Policy on Management of Environmental Data and
Environmental Science Information. NOAA Directives Manual
16-11. December 9, 1971.
-------
Appendix B
Presentations to the Committee
1. Ms, Patricia Berger, Chief, Library Systems Branch, 0PM
2. Mr. Kerrigan dough, Special Assistant to the Administrator
3. Dr. Homer Hall, Information Analysis Project Director,
Rutgers University, Consultant to the Committee
4, Dr. Steve Heller, Office of Chemical Information, OTS
5. Dr. Kent Hughes, Deputy Director, National Oceanographic
Data Center, NQAA . .
6. Mr. John Krobock, Associate Professor Lehigh University,
Consultant to the Committee
7. Mr. Calvin Lawrence, Deputy Director, Environmental Research
Information Center, ORD
8, Ms. Marion Mlay, Director, Program Evaluation Division, GPM
9. Mr. Les Needles, Vice President, Sigma Data Computing Corp,
10. Mr. Neil Ruzic, Island for Science, Inc., Consultant
to the Committee
11. Dr, Sidney Siegel, Senior Science Advisor, OTS
12. Ms. Libby Smith, Chief Librarian, EPA,
Research Triangle Park, N.C.
13. Mr* Morris Yaguda, Chief, Information Planning &
Coordination Branch,
-------
197S
1) C*t, cnzzvit lit. It
^ <**»•• r-y Mrwii^
2} £*t. OOflfe »£
3} Ixvdp* «ff«
of MrriM* to M*t
(Xllou »Tt to ««lu*«*
et
(311
ifn«d d»
l of 4.1
tr. K
S* ar ch Sy atBtfM of gyA
finding* t
1) Juulyn tb«
of BJk
Z!
Mz
Info, o«Dt*r (NRK)
9.A.
H.A.
r- Agency
to M. , EPA
1979
1> ft.
wiitn«itt»H
pro-
* {70
- 13 D
f 57 Div.
OX OtBAT QflW BMIA*
9«r», 15 MM Staff
If w*.
EPA
100
llz t »«t*t
H.A,
QAA
K.A.
U.A.
-------
Bfce*rt*i»ey (bent ncwztaisCf «bouc Too Hurt Too auda, ourriil^jiliTy L*dt lf£*esiv» L.e, *
info, ii ilhui lafa, eu Info. » Beqfalc » of Is* Hftiit, ngt. Ot»ctien
!
Cotla incltrf. «ll
iamtf ici*nt XaairlHif v of
(••g* 6 balding*,
i«t», li»it*d iy
of
(—— M«z MIUWWH ncl** *id*ly, All Uxm Oeald
of md*lvJij» of
CcllcctwS MnieoriBf a*bi m&
-------
Licit 9< Ifc.Ai.ly *«il~
l*i>«ri«a
•» *»
KPA'I 5P
(COuid Be
-------
* «w* •£**«*«* Srtafiliab * Wat to ***i*« C»r,lop Hoc*
Pec Kwpiiis Vwpl* to Or\f«aii« ud «pp«
-------
sf JtttabUih
for KMBiJUM »•«»:** to
Coat Mflcf.^t* Utcrvtun Cstafeam
taliey ......... _ CutrtliCM
-------
S.ITCJI mqnlraaiita fox crou
C Itili
-------
Iat«fr«t,»a
------- |