| II.
MAJOR SUBSTANTIVE RECOMMENDATIONS OF THE WORKING
PARTY |
| 15. |
The remainder of this document
highlights the substantive recommendations that
emerged from the meetings of the Working Party
and the Workshop held in October 1999. |
|
|
| A.
Major Recommendations of the First Meeting of
the Working Party, held in Bangkok from 24 to
26 September 1997 |
| A.1. |
Noting that the external financial
assistance for developing countries for the
2000 round of population censuses was likely
to be at a reduced level compared to the past,
the Working Party recommended that cooperation
between countries in planning and conducting
census and related activities should be enhanced. |
| A.2. |
Recognizing that the parent
organizations of some of its members were engaged
in a wide range of evaluations of technology
options for various aspects of population data
processing, the Working Party recommended that
the results of such evaluations should be shared. |
| A.3. |
The Working Party recommended
that a careful assessment should be made of
all available options in meeting equipment requirements
for census operations, including rental of equipment
and outsourcing of some of the tasks. |
| A.4. |
The Working Party identified three core areas
as the major focus of the project, as follows:
- GIS in census and survey
applications
- Imaging and data capture
technology
- Application of Internet
based technology
|
| A.5. |
As the Internet could be used
at various stages of population data collection
and dissemination, the Working Party decided
to focus its second session on the topic and
related issues. |
| A.6. |
In light of its major focus the Working Party
decided that the topics and the sites of the
three pilot applications to be undertaken
should be as follows:
- Imaging technology - Indonesia
- Use of GIS for census
operations and dissemination - Philippines
- Use of GPS for preparation
of census enumeration area maps - Bangladesh
|
| A.7. |
In view of New Zealand's expertise
in several relevant technological areas, the
Working Party recommended that it should be
invited to nominate an expert to serve as a
member of the Working Party. |
| A.8. |
The Working Party recommended
that a sub-group should be formed to review
the imaging technologies and facilitate their
selection and deployment. It should include
Australia, Japan, New Zealand and Singapore,
with Indonesia as its convener. In addition
the United Nations Statistics Division (UNSD)
should be invited to participate. |
| A.9. |
The Working Party recommended
that the secretariat should conduct a survey
to assess the use of information technology
in population data collection, processing and
dissemination. The results of the survey would
be submitted to the ESCAP Committee on Statistics
in 1998, after a review by the Working Party. |
| A.10. |
The Working Party recommended that five issues
of the project Newsletter be produced with
the following focus:
- Introduction to the project
- proceedings of the 1st meeting of the
Working Party
- Internet
- GIS (including GPS and
satellite imagery)
- Imaging, data capture
and hand-held devices
- Summary and evaluation
of the project
|
| A.11. |
The Working Party recommended
that the following three sets of guidelines
be produced under the project as follows:
|
Topic |
Coordinator |
| a. |
Dissemination |
Secretariat3/ |
| b. |
Mapping |
Bangladesh/Philippines |
| c. |
Data entry, capture,
processing and archiving |
Indonesia |
3/
Subsequently, New Zealand volunteered and was
designated as coordinator for the guidelines
on population data dissemination.
|
| A.12. |
The Working Party recommended
that each set of guidelines should draw from
technology practices in the countries and provide
options for their implementation. |
| A.13. |
The Working Party recommended the following
provisional timetable for its future meetings
to be held at suitable venues:
- second meeting - as early
as possible but not later than early April
1998
- third meeting - September
1998
- fourth meeting - first
half of 1999
|
| A.14. |
The Working Party noted that
further meetings might be possible if savings
anticipated in the project materialize. |
|
|
| B.
Major Recommendations of the Second Meeting
of the Working Party, held in Singapore from
1 to 3 April 1998 |
| B.1. |
Expressing appreciation to
New Zealand of its acceptance to become a member
of the Working Party, the Working Party amended
its "Mode of Functioning" to reflect the enlargement
of its membership. |
| B.2. |
The Working Party requested
the secretariat to dedicate a section to the
Working Party at the Statistics Division's website,
featuring the newsletter, meeting documents,
recommendations, guidelines and other work done
in the project. |
| B.3. |
Welcoming the recruitment
of a programme assistant for the project, the
Working Party requested the secretariat to start
collecting links to sites and documents concerning
IT applications to population data. |
| B.4. |
It also asked the secretariat
to investigate the possibility of establishing
at its website a repository of documents on
population data technologies. |
|
|
| Recommendations
on the use of Internet |
| B.5. |
Noting the positive experiences
of Singapore and other countries in using the
Internet in data dissemination, the Working
Party encouraged in general all statistical
offices to establish an effective Internet presence,
and use it innovatively in census and survey
operations. |
| B.6. |
The Working Party noted the
innovative efforts of some countries in using
the Internet for data collection and encouraged
them to share their experiences with other countries. |
| B.7. |
The Working Party recommended
the NSOs to adopt an incremental and modular
approach for developing a website, especially
when resources were limited. |
| B.8. |
Noting that several advanced
NSOs were depending on e-mail for their daily
communications which had significantly increased
their productivity, the Working Party recommended
an extensive use of e-mail and other Internet
applications by all NSOs in census planning,
document sharing, data transfer and other related
functions. |
| B.9. |
The Working Party noted that
to achieve maximum benefits from the new technologies,
changes might be necessary, among other things,
in organizational structures, coordination arrangements
and individual responsibilities. It recommended
that NSOs should take a critical look at the
population data operations as a whole with a
view to reengineering them to meet the requirements
and opportunities of Internet technologies.
It stressed that systematic training of staff
was essential to instill a holistic approach
that took the technology into account from the
beginning. |
| B.10. |
The Working Party recommended
that metadata be included as an essential component
at census websites as they added considerably
to the value of population data. The metadata
could include such information as the general
organization of the census, coverage, census
forms, instructions to field workers, methods
followed in derivation of numeric results, accuracy,
sampling methodology, and references to published
reports. |
| B.11. |
Recognizing that there was
a glut of information competing for users' attention,
the Working Party felt that it was necessary
to promote statistical websites by using all
available means, including announcements on
national and global search engines. |
| B.12. |
Discussing other ways to increase
the visibility of population data, the Working
Party recommended equipping websites with a
search facility, preferably the types allowing
open text searches; it further recommended systematic
use of HTML metakeys and descriptive titles
that were automatically picked up by referencing
engines and crawlers. |
| B.13. |
Mindful that a large number
of Internet users in developing countries were
behind a narrow bandwidth connection, the Working
Party cautioned authors not to include large
graphics components on pages that made download
times prohibitive. The Working Party emphasized
that it was important for the developers to
test web pages with low-end browsers that did
not support graphics, and provided a text-only
alternative. |
| B.14. |
The Working Party noted that
the development of electronic commerce technologies
and secure data encryption had opened up new
possibilities for commercial statistical products.
It urged NSOs to consider that new area as another
means of data dissemination. |
| B.15. |
Noting the benefits of the
Internet in making statistics available to users
faster and in greater depth than previously,
the Working Party recommended that statistical
offices streamline their release approval/authorization
processes to match the significant progress
achieved in the technical dissemination process. |
| B.16. |
The Working Party noted that
a number of software, applications, and solutions
had been developed by NSOs which could be shared
among the countries/areas of the region. It
recommended that such software and applications
which might be available for distribution be
listed on the ESCAP homepage. The Working Party
expressed its gratitude to the Singapore Department
of Statistics for identifying FASTAB as a product
which could be shared in the region. |
| B.17. |
Recognizing the importance
of the influence of policy and decision makers
in prioritization for budget allocations, the
Working Party recommended that an awareness
package should be prepared and that its content
should stress the importance and purpose of
census taking including the necessity of making
available census data quickly go the users.
The content of the package could focus on strategies
and application of technology in the various
phases of census operations with a focus on
speed processing and timely dissemination of
census data. The package could be in the form
of a short video with supporting documents,
which should be made available for the 2000
round of censuses. |
|
|
C.
Major Recommendations of the Third Meeting of
the Working Party,
held in Bali from 7 to 9 January 1999 |
| C.1. |
In view of the recommendation
of the mid-term review of the project that a
workshop be convened in October 1999, the Working
Party stressed the importance of producing outputs
well ahead of the workshop. |
| C.2. |
In reviewing and revising
the outlines of the three guidelines in the
light of its previous discussions on their scope
and the pressing time constraint, the Working
Party recommended that the titles of the three
guidelines be as follows:
- "Guidelines on
the application of new information technology
to population data dissemination"
- "Guidelines on
the application of modern mapping and GIS
technologies in census operations"
- "Guidelines on
the application of new technology to population
data collection and capture"
|
| C.3. |
The Working Party agreed that technology
sections of all three guidelines should deal
with the following issues, for which the material
presented in the Working Party meetings offered
a good starting point:
- Explain concepts necessary to understand
the technology and the objectives of applying
it
- Cover requirements for planning and training
to put the technology in place and maintain
it; cover also issues related to daily operation
and management
- Contain advice for estimating total costs
involved in applying the technology
- Contain consolidated good country practices
in introducing and applying the technology.
|
| C.4. |
In reviewing the results of
the survey conducted by the secretariat on application
of IT to population data, the Working Party
noted that it provided valuable information
and an useful analysis of the state of IT application
in the region. The secretariat document revealed
a wide gap between developed and developing
countries in applying information technology,
and again confirmed the importance of achieving
the project objectives. The Working Party recommended
that the three guidelines to be prepared under
the project should take into account the survey
results and focus on the apparent needs of the
developing countries. Nevertheless, the considerable
progress made by NSOs in applying IT in their
latest censuses and surveys allowed the Working
Party to take an optimistic view that developing
countries were able to seize many opportunities
offered by new IT. |
| C.5. |
The Working Party praised
the quality of the contributed papers made available
to the meeting and noted that they formed a
significant repository of applied technology
information that should be made available in
formats required by the developing countries. |
| C.6. |
The Working Party recommended
that an item be introduced on the agenda of
the eleventh session of the Working Group of
Statistical Experts to discuss the outcome of
the activities of the Working Party. |
| C.7. |
The Working Party welcomed
the introduction by the projects's Mid-Term
Review of a training workshop, tentatively in
October 1999, as a channel to make the project
outputs available to a wider audience, and recommended
that:
- The broad objective
of the Workshop would be to sensitize participants
to the opportunities that modern information
technology provided in population data operations
and to improve the guidelines produced under
the project.
- It should be targeted
at IT and statistical managers who had influence
on the selection of technologies for census
and survey operations
- apart from conventional
lectures, the participants should be given
hands-on opportunities and time to test
and study new applications
- Working Party members
should act as resource persons in the workshop
and would be responsible for organizing
vendor demonstrations in areas they worked
on
- the guidelines and
pilot applications be used among the training
material for the Workshop
- the workshop should
be held in a venue where participants had
new technology readily accessible
- the secretariat use
its own knowledge about the countries and
the results of the survey to choose participating
countries
- the resource persons
of the workshop should take into account
the survey results in preparing their inputs
and presentations
|
| C.8. |
The Working Party agreed that
the participants of the workshop would benefit
from product demonstrations by representatives
of hardware and software providers, but noted
that such opportunities should not be restricted
to particular organizations. |
| C.9. |
The Working Party noted the
observation that the issue of developing an
awareness/training package for promoting effective
and efficient utilization of IT in population
censuses and surveys needed reconsideration.
It proposed that SIAP should reorient the aim
of the package for the purposes of training
NSOs in new technology applications. The material
developed would be based on the outputs of the
project, among others, and be used in current
and future regional and national training courses. |
| C.10. |
The Working Party recommended
that the project outputs should also be made
available to a wider audience by disseminating
them on CD-ROM, and be used in national as well
as SIAP's institutional and outreach training
programmes. |
|
|
| D.
Major Recommendations of the Fourth Meeting
of the Working Party, held in Manila from 7
to 9 July 1999 |
| D.1. |
The Working Party agreed that
the statistical offices were not the major mapping
agencies in the countries and therefore did
not have primary responsibility to produce digital
maps. However, the use of maps for statistical
purposes made it necessary for the statistical
offices to play an active role in coordinating
efforts to produce digitized maps where they
were not available. The meeting noted that Bangladesh
Bureau of Statistics, through its project, had
in fact been instrumental in bringing together
the major agencies of the government to coordinate
efforts to produce maps in digital formats and
to use the same for GIS applications. |
| D.2. |
The Working Party emphasized
the importance of having all guidelines in their
final draft formats before the Workshop, which
was to be held in October 1999. It noted that
the Workshop would be used as a test audience
for the guidelines and that they were likely
to require some level of editing after the Workshop. |
| D.3. |
The Working Party made specific
observations and recommendations on the three
guidelines. It agreed that it was not in a position
to recommend any of the covered technologies
prescriptively or to endorse any particular
brand name. It accepted that the guidelines
would not be official United Nations publications,
and that they would still be subject to the
standard United Nations disclaimers. |
| D.4. |
The Working Party confirmed
that the content requirements that had been
set for the guidelines in the previous meeting
were still valid. |
| D.5. |
The Working Party also confirmed
that the guidelines needed to contain advice
for estimating total costs involved in applying
the technology. It acknowledged, however, that
it was difficult in some cases to indicate exact
monetary costs for applications, especially
if they had been developed over a period of
time in-house involving an institutional learning
process. The Working Party recommended the inclusion
in the guidelines, wherever applicable, of alternative
ways of acquiring cited technologies, including
the use of public domain software, purchase
and leasing of equipment and software, in-house
development, and partial or full outsourcing
of related tasks. |
|
|
| Recommendations
on pilot applications |
| D.6. |
The Working Party expressed
its satisfaction with the progress made on the
three pilot studies and stressed that their
reports should be completed before the end of
September 1999. However, to receive comments
from the members of the Working Party, the draft
manuscript should be circulated via e-mail by
15 September 1999. |
| D.7. |
In discussing the formats
of the reports of the pilot projects, the Working
Party recommended that each report should include
the statement of the problem, situate the problem
within the main activity that it was part of,
place the problem in the sequence of events,
provide details on the approach and methodology
adopted, describe the experience, describe measures
adopted for solving the problems that were encountered,
and report on lessons learned. It was informed
that in each pilot project document the deliverable
outputs were identified and that the recommended
common features of the report format were consistent
with them. |
| D.8. |
The Working Party recommended
that to the extent possible, the experience
of the pilot projects be incorporated into the
guidelines to be produced under its guidance. |
| D.9. |
The Working Party decided
that the Workshop, scheduled from 12-20 October
1999, should utilize papers produced for the
Working Party meetings, experiences collected
in the three pilot applications developed, and
the three guidelines as core material for the
Workshop. To make a distinction that the Workshop
had a limited scope as opposed to general IT
training, the Working Party decided to name
it as "Workshop on application of new information
technology to population data". Notwithstanding,
the Working Party agreed that it was paramount
to include in the Workshop sections that facilitated
the understanding of the overall role of new
technology in conducting censuses and surveys. |
| D.10. |
The Working Party decided
on the outline for the Workshop and appointed
organizers and allocated time for each module. |
| D.11. |
The Working Party agreed that
some sessions could benefit from relevant product
demonstrations and asked members to extend their
expertise to assist the organizers and the secretariat
in proposing and choosing the candidates for
them. |
| D.12. |
The Working Party discussed
the development of an awareness package on the
application of modern technology to population
data. The Statistical Institute for Asia and
the Pacific (SIAP) had been designated to develop
the package in close cooperation with the Statistics
Division of ESCAP and the members of the Working
Party. |
| D.13. |
The Working Party agreed on
the proposed texts concerning the problem definition,
the objective and the target group for the package.
The suggestion was made to add the point that
the decision-makers had to be aware of the fast
moving techniques and should take measures to
stay up-to-date. |
| D.14. |
The Working Party decided
that the awareness package should have the form
of a multimedia-presentation which could be
disseminated via Internet, video tape, CD-ROM
or diskette. The three guidelines might, after
their completion, be transferred (in HTML- or
PDF-format) to a CD-ROM, with inclusion of navigation
functions for better readability. |
| D.15. |
The Working Party recommended
that only one CD-ROM presentation be released
by the ESCAP secretariat, containing the various
outputs of the project, such as guidelines,
and the awareness package prepared by SIAP. |
|
|
| E.
Recommendations of the Workshop on Application
of New Information Technology to Population
Data, held in Bangkok from 12 to 20 October
1999 |
|
|
| General,
IT management |
| E.1. |
The Workshop agreed that the
conduct of censuses and surveys was necessarily
becoming increasingly technology intensive.
It recommended that national statistical offices
keep abreast of the latest information technology
by continuously monitoring technology evolution
and by upgrading production and office systems
periodically. |
| E.2. |
Appreciating the excellent
cooperation and contribution received during
the project, the Workshop recommended that technologically
advanced offices continue to share with others
their experiences in adopting new information
technologies. |
| E.3. |
Noting that modern data capture
technologies (OMR, OCR/ICR, CAPI, CATI, Internet
data collection) had uses in many sectors, the
Workshop recommended that in order to keep IT
applications cost-effective, census and survey
organizations should collaborate, among themselves
and with other agencies, in the procurement
and post-census use of the equipment and software. |
| E.4. |
The Workshop noted that for
many countries budgetary constraints hampered
the effective application of new technology.
It requested the bilateral and multilateral
donor agencies to increase their assistance
to developing countries for IT applications,
and recommended that the technical cooperation
among developing countries (TCDC) modality be
promoted for an enhanced sharing of IT experience
and skills through expert visits and study tours. |
| E.5. |
The Workshop recommended that
statistical offices upgrade their organizational
IT knowledge and create a modern IT culture,
and develop prudent procurement methods to match
the skilful and articulate marketing techniques
of private sector vendors. |
| E.6. |
The Workshop recommended that
Governments should take into account in their
procurement rules the overall costs and benefits
that each technology alternative offered in
the long term, and not take decisions solely
on the bid price for a particular application. |
| E.7. |
The Workshop emphasized that
it was crucial for senior management in the
national statistical offices to increase its
awareness of trends in information technology
and the associated costs and benefits, and to
improve related management skills. |
| E.8. |
The Workshop recommended that
national statistical offices should ensure that
any vendor being considered for the supply of
new technology systems was able to substantiate
its claims. Statistical offices should have
a benchmark drawn up addressing their requirements,
before the commencement of the evaluation process.
They should also ensure that staff evaluating
potential systems and vendors have a good knowledge
of the requirements and of the technology being
evaluated. |
| E.9. |
The Workshop recommended that
NSOs and census organizations make full use
of the opportunities that new information technology
offered in the conduct of censuses. They should
bear in mind that no stage of a census could
now be planned and executed without taking technology
into account and that new technology had merged
certain stages in the census operation. The
Workshop recommended that census organizations
make corresponding changes in their organizational
and management structures, and adjust the resources
available for IT procurement, recruitment of
skilled staff, and training of existing staff. |
| E.10. |
Given that statistical offices
had to take the whole range of census operations
into consideration while assessing the implementation
of new technology applications, the Workshop
recommended the application of quality management
strategies as a useful method for control of
the whole process. Further, the interoperability
of the various components to be chosen required
special attention, not only with regard to the
operational aspects but also in terms of the
integrity of the huge masses of data to be processed. |
| E.11. |
Recognizing that many developing
countries were using public domain software
packages, the Workshop recommended that ESCAP
should promote sharing of experiences on the
use of such packages with a view to maximizing
the benefits of those applications. |
| E.12. |
The Workshop recommended that
statistical offices should avoid procuring hardware
and software that did not run under common operating
systems, that did not provide integration with
other systems, that was not easily extendable,
that had no indication of long-term support
and that was likely to lead to dependency on
one vendor. |
| E.13. |
Noting that electronic format
had many advantages over hard copy format, the
Workshop recommended that statistical offices
should aim at digitizing census and survey information
as early as possible. That would involve greater
utilization of existing electronic records (administrative
records), adoption of computer-aided interview
technologies, and scanning of census forms immediately
after enumeration. Electronic format minimized
manual handling of forms and allow maximum flexibility
in data verification and editing. |
| E.14. |
The Workshop noted that it
was essential for statistical offices to ensure,
as part of the evaluation process, that selected
vendors have the commitment and capacity to
train the statistical office staff in the hardware
or software, and to provide continuing service
and support. |
| E.15. |
Considering scarce resources,
especially in the small developing countries
and areas, the Workshop recommended that on
a subregional level governments should find
ways to cooperate in the purchase and utilization
of expensive current technology, e.g., by sharing
the cost of acquisition and responsibility for
operation and maintenance of such equipment.
Further, the Workshop recommended that governments
of developed countries and areas, which operate
such advanced-technology systems, should make
their use available to developing countries
in the region, preferably at nominal or no cost. |
| E.16. |
The Workshop noted that language
capabilities of data capture and dissemination
software were important in many countries. In
the area of data capture, the OCR/ICR engines
achieved very high recognition rates for hand-written
characters in a limited number of languages;
that efficiency was not matched for numerous
other languages in Asia. Similarly, many NSOs
required bi- or multilingual capabilities for
data tabulation and dissemination software.
The Workshop recommended that the NSOs express
language capability as one of the prerequisites
for software acquisition, and recommended that
the software developers expend efforts in incorporating
local language and multilingual capabilities
in their products. In that regard, it was noted
that the Workshop had provided an excellent
opportunity for the vendors to better understand
the needs of the NSOs and also explain some
of the features of their products which were
of interest to the NSOs. |
| E.17. |
The Workshop recommended that
further technical meetings be held after the
2000 and 2001 censuses to share information
on technology lessons learned, and to promote
effective data utilization and dissemination. |
| E.18. |
To facilitate exchange of
experiences, ideas and information on resourcing
and other topics, it was recommended that an
e-mail based discussion group be established. |
|
|
| Data collection
and capture |
| E.19. |
As the current data capture
technology provided increasingly powerful means
of handling data on numerous topics for large
collections, the pressure for expanding the
scope of the census was mounting. The Workshop
cautioned that in considering those demands,
census statisticians must not ignore the operational
aspects of actual data collection in the field,
the skill levels required for data collection
and handling, and the technical requirements. |
| E.20. |
The application of IT would
also assist countries in improving the management
of errors and coding of captured information
from censuses and surveys. The Workshop recommended
that greater sharing of information should be
promoted in those areas, including computer-assisted
coding. |
| E.21. |
The Workshop recognized that
selection of data capture technology was a crucial
success factor in census taking. It advised
census organizations to assess carefully all
costs, including the implications for various
census operations, involved in the selection,
procurement, operation, maintenance and management
of capture technology. |
| E.22. |
The Workshop recommended the
conduct of at least one and preferably two major
tests using real forms, real enumerators and
real respondents to test systems. Testing was
needed for:
- the selection of the
preferred technology
- refinement and improvements
in the technology
- development of procedures
and arrangements related to the implementation
of the technology
- the building of awareness
within management about how the new technology
should be handled
- calculation of the
resources needed for the main event
- preparation of the
content, schedule, and methodology of training
to be carried out.
|
| E.23. |
The Workshop recommended that
census organizations make full use of the flexibility
that was offered by new imaging and recognition
technologies, for instance by planning for an
early release of results for the most important
topics. |
| E.24. |
The Workshop recommended that
census organizations evaluate data capture solutions
carefully taking into account country circumstances.
Evaluation results obtained elsewhere were not
necessarily directly applicable, due to differences
in handwriting patterns, questionnaire design,
and availability of quality paper, ink and printing
facilities. It noted that competitive benchmark
testing had become a standard evaluation method
in large census organizations all over the world. |
| E.25. |
Noting that the available
character recognition software was developed
for universal use and that the turn-key OCR/ICR
solutions were restricted to data capture (and
did not cover the whole census operation), the
Workshop recommended that software developers
incorporate in character recognition applications
statistical features, such as classifications
that assisted in data coding. |
| E.26. |
The Workshop recommended that
statistical organizations planning to use OCR/ICR
should develop procedures to control the quality
of recognition. It was particularly important
to search and check for non-random bias caused
by systematic recognition errors. |
| E.27. |
The Workshop agreed that imaging
should not be used simply as a data capture
replacement technology, and recommended that
statistical organizations identify which other
census processes were affected and determine
how they could be made more efficient and cost-effective. |
| E.28. |
The Workshop recommended that
census and survey offices should consider outsourcing
as an option for implementing elements of censuses
and surveys. It noted that the feasibility of
outsourcing depended on national circumstances,
the organization's own resources and skills,
and the availability of external partners. It
heard of the experiences of the Singapore Department
of Statistics in developing an innovative multi-mode
data capture system for the year 2000 census
by using several external developers. The Workshop
noted that the multi-vendor approach required
clear delineation of responsibilities for system
development and support, which could be conveniently
achieved by using a prime contractor approach. |
|
