|
Symposium
2001/34
13 July
2001
English
only
|
Symposium on Global Review of 2000 Round of
Population
and Housing Censuses:
Mid-Decade
Assessment and Future Prospects
Statistics
Division
Department
of Economic and Social Affairs
United
Nations Secretariat
New
York, 7-10 August 2001
Statement from Islamic Republic of Iran *
Abbas-Ali Zali **
CONTENTS
B. Strategies for involving
stakeholders in census activities
D. Adapting new technologies
to census operations
E. Maintaining census-related
activities during the intercensal years
F. Identifying and resolving
problems of census mapping
G. Post-enumeration surveys:
are they worth it or not?
1.
The
fifth census of the Iranian decennial national census of population and housing
took place from 23 October to 16 November 1996. Compared with the previous
ones, the 1996 census enjoyed some advantages, as follows:
1.
To
enrich the national database, the identity card number and serial number were
included among the items of the census questionnaire;
2.
To
complete the allocation of 10-digit postal codes in cities, towns and villages,
and to expedite the assigning of the national identity number to all nationals,
the place-based listing forms of the post office were used in the census;
3.
To
give equal training to everybody involved, an educational video cassette was
prepared to teach the forms and instructions;
4.
To
capture data from the filled-out questionnaires, the optical mark reader (OMR)
method was used for the first time, which expedited the job;
5.
The
census results were disseminated as printed materials and on CDs just six
months after the field stage;
6.
The
latest version of the International Standard Classification of Occupations (
ISCO-88 ) was used for the first time;
7.
Also,
the imputation method was applied for some non-response data, which contributed
to an early data dissemination. The reasonable relationships between different
items of the questionnaire were determined and errors were diagnosed; then the
incorrect data were replaced with the correct data, which was obtained after
considering indices and expected logical relations.
2.
In
the Islamic Republic of Iran, the Statistical Centre of Iran is by law
responsible for implementation of the censuses. Every census is launched by the
President’s decree. Since there is not a permanent and continuously working
organization for carrying out the censuses, a headquarters is established
consisting of a group of managers and senior experts from the Statistical
Centre of Iran (SCI) and the representatives of the interested organizations
for each census. The people gathered are then arranged in different groups and
committees. The headquarters of the 1996 census was formed in the fall of 1994,
two years prior to the field operation, and began work by discussing the census
objectives, reviewing former censuses and recognizing the statistical topics.
As the discussions began, the significance of considering the data users’
notions in all stages of the job, particularly in deciding on the census
content—the topics and concepts and definitions—was deeply felt.
Accordingly the following
organizations were called on:
·
Nomads
Affairs Organization
·
Islamic
Republic of Iran Broadcasting
·
State
Welfare Organization
·
Central
Bank of Iran
·
Tehran
City Hall
·
Ministry
of Labour and Social Affairs
·
Ministry
of Education
·
Ministry
of Health
·
Social
Security Organization
·
Ministry
of Culture and Higher Education
·
Ministry
of Housing and Urban Development
·
Plan
and Budget Organization
·
Ministry
of the Interior
·
Ministry
of Jahad-e-Sazandegi
·
Plan
and Budget Organizations of Ostans ( provinces)
·
Ministry
of Cooperatives
C. Strategies for choosing
among data-collection methods as
sources of demographic and social statistics
3.
Proposals
presented by the organizations and persons answering the call mostly included
their intended items and, in some cases, presentation and modification of
definitions and concepts as well as frames of results and tables. Altogether,
110 proposals were offered on the census items. In the first step, these items
were classified at headquarters into three categories: general use, special use
and non-statistical use.
4.
The
first two categories were compiled from three methods: census, registers and
sample surveys. Investigation, classification and finally selection of the
proposed topics were made according to the following criteria.
·
Recognition
of main characteristics of the
population, household and housing within the national boundaries so as to fill
the existing statistical gaps as well as meet the current statistical needs for
policy and decision making;
·
Necessity
of response to national priorities and fulfilment of the development plan needs
for general information as well as information on specific groups of the
population;
·
Capability
of representing results and comparing them at any geographical level and on the
general and basic items with the results obtained from the previous censuses;
·
Giving
priority to the basic information;
·
Achievement
of international and regional comparability while giving full consideration to
national needs; and
·
Enjoying
practical efficiency and reasonable costs during the implementation of census
plans.
5.
Moreover,
other factors had to be taken into account, such as determination of the
priorities of the census items, volume of the job, and administrative and
operational limitations of the census, as well as the nature of the census,
involving a large number of enumerators with average skills and capabilities.
6.
It
was necessary to prepare a plan while considering all of the above-mentioned
principles and criteria. Several seminars and meetings were held at census
headquarters with related groups and committees. Also present were
representatives of the concerned organizations, researchers and lecturers at
the universities. Each of the new statistical items was evaluated according to
the intended criteria. During the meetings, discussions were held on
preparation of definitions and concepts, needed classifications for some topics
and items confirmed to be included in the census. Also discussed in
collaboration with the concerned organizations was the process of presentation
and dissemination of the census results. It was not necessary to persuade the
concerned institutions of the reasons that some of their requested topics were
not included in the census.
7.
In
general, the primary aim of the population and housing census in every country
is to meet national needs in the area of specific demographic and housing characteristics.
In some instances, when there is a discrepancy between national needs and
international recommendations, the national needs should be given priority.
8.
Additionally,
in these meetings, international advisors and overseas university professors
were consulted. Census plans available in some other countries—including
Australia, Canada, China, France, India, Japan, Malaysia, Pakistan and the
United States—were investigated.
1. Topics
considered for the 1996 census
9.
Some
items discussed in these meetings, and the reasons for acceptance or rejection
of them in the 1996 population and housing census are presented below.
a. Cause of migration
10.
This
topic was not included in the census agenda for the following reasons, but it
was to be included in the sample surveys:
·
Cause
of migration does not play a significant role in the population projections;
·
Cause
of migration can change over time; therefore, it lacks stability;
·
In
some instances, more than one factor is involved in the cause of migration; thus,
it is difficult to determine the main reason for migration.
b. Information
on fertility and mortality
11.
Collection
of information on fertility and mortality is a time-consuming job and requires
well-trained enumerators. Therefore, such data were to be collected in the
sample surveys. After the end of census taking, a sample survey and a
post-enumeration survey were immediately executed concurrently.
c. Enumeration
of building instead of place of residence
12.
This
topic was not approved owing to the complexity of the building definition and
the additional burden on the enumerators.
d. Means of
commuting
13.
This
topic was first approved and included in the pilot census questionnaire.
However, considering the pilot results obtained, and the fact that the traffic
problems exist only in some large cities, inclusion of this topic in the main
census questionnaire involving all areas of the nation was not cost-effective.
e. Mobile
nomadic households
14.
The topic of mobile nomadic households and how to
satisfy their needs covers only a small part of the country. Therefore, it
should be investigated in the sample surveys. In the 1996 population and
housing census, only basic questions were included in the questionnaire.
f. Radio and television coverage in villages
15.
Because of the needs of the Islamic Republic of Iran
Broadcasting for this kind of information and lack of access to this
information through other methods, questions on radio and television coverage
were included in the 1996 census questionnaire.
g. Cause and type of physical and mental disabilities
16.
Considering the needs of the State Welfare
Organization for basic information, and international recommendations in this
regard, the question on households with disabled persons was discussed. After
much consideration, this topic was approved, but questions on the cause of the
disability and other details were left for sample studies.
17.
After summing up the main results of the comments,
and development of the preparatory census plan in the autumn of 1995 (that is,
a year following the formation of the census headquarters, and a year prior to
the 1996 population and housing census), the primary census plan was tested.
18.
It is necessary to mention that when the census plan
was being developed in the headquarters over a two-year period, some ad hoc
tests were employed on some items in
addition to the above test, and the probable inclusion of them in the
questionnaire was discussed. The results obtained were very useful in the final
development of the census plan.
19.
After conducting several tests and evaluating the
results, discussions were held with the representatives of organizations,
executive bodies of provinces, professors of the universities and researchers
in different seminars and meetings. Following these discussions, the final
selection of topics and finalization of the census plan began.
20.
In general, there are many problems in collecting
information, among them the speed of collecting information and the accuracy of
the data collected. By developing new technology, traditional methods are being
abolished and new methods are being applied to census operations. Completing
paper questionnaires is a very time-consuming operation owing to the
question-and-answer process and investigation of incorrect data.
21.
Because of the high volume of data in censuses, it
might be impossible to employ the above-cited methods. Making the society aware
of the importance of answering questionnaires is an issue which has been taken
into account by advanced countries for many years. Results obtained have been
successful. Therefore, the data collection is carried out without the aid of
the interviewer in the censuses of such countries. However, what is important
is the stage following completing the questionnaires—that is, what kind of
methods are employed for data capture. They include traditional (keyboard) data
entry, optical mark reading (OMR) and optical character recognition (OCR),
depending on the type of the enumeration and the expenditure. Unfortunately,
OCR is not commonly used in countries such as the Islamic Republic of Iran due
to the local languages. However, despite the problems existing in the
employment of OMR, the use of this method in the 1996 census decreased the data
dissemination period, for the first time, to less than six months since the
implementation of the census.
22.
In the traditional (keyboard) data-entry method,
ready-made packages are used for designing questionnaires. However, the use of
the packages in some questionnaires is impossible for certain reasons.
Therefore, programming is called for.
1. Data processing
23.
After collection of data, it is necessary to
edit, correct and process data generated in the census. The traditional method
commonly uses conventional files. After collection and sorting of data in the
cited files, the collected data are investigated. This is carried out through
traditional programming and using ready-made packages. The data-capture
operation is the next stage, and this is done by using traditional programming
or software packages such as Integrated Microcomputer Processing System (IMPS).
24.
As
mentioned before, the SCI usually establishes a provisional preparatory
headquarters two years before the census. It comprises the experienced experts
of the Centre who are responsible for the planning and administration of the
census. The headquarters, by taking advantage of its different groups and
committees, manages the pre-census and post-census activities, as well as the
actual census.
25.
These
activities consist of selection of topics, questionnaire preparation,
cartographic work, planning for training and control, census publicity, staff recruitment,
budgeting, and the like for pre-census training; fieldwork, control, and report
on the progress of the work during the census; and finally the post-enumeration
survey, preliminary review of the results, data entry and data processing, data
capture, tabulation and dissemination after the census. Along with the above
activities and as a part of them, a committee is organized for documentation
and recording. Its main responsibility is to collect and classify all the
documents related to various stages of the census and store them for future
utilization by the Centre experts and other users.
26.
After
the end of the census, the documentation group as well as other census
divisions will be dissolved and its members will return to their previous
duties in the Centre. All documents
prepared by the group are kept in the library of the Centre. Census information
is released after the data processing as publications in various geographical
regions for public use.
27.
After
the completion of the census by the headquarters and its related groups and
committees, all the related activities concerning the census, such as ad hoc
studies using census information, are performed through the departments of the
Statistical Centre of Iran which are concerned with socio-economic statistics.
28.
Major
problems emerged in the mapping operations for the Iranian 1996 census, as
described below:
·
The
first problem concerned base maps. Although the Statistical Centre of Iran is not
legally in charge of developing maps at the national level, due to the
difficulties and inefficiencies concerned with developing the base maps, the
Centre is forced to produce, correct and update the maps required for
implementation of censuses and surveys. In each case, however, some of the maps
are provided by the National Cartography Centre, and the Statistical Centre
merely adapts them to the project’s framework.
·
There
was a lack of adequate facilities, technical and financial support and
expertise needed for developing the maps and updating them at different stages
of surveys and censuses.
·
The
mapping project was unable to attracting professionals at the national and
subnational levels because of recruitment restrictions.
·
The
lack of permanent geographic units resulted in the inability to compare
collections of statistical data in different time series. In the 1996 census,
delays were encountered in determining enumeration districts.
·
Frequent
changes in administrative divisions caused some problems and imposed a heavy
burden on updating the maps accordingly.
·
There
is no accurate base map for developing a Geographic Information System (GIS),
which could greatly contribute to the removal of problems concerned with
updating and making use of maps in censuses and other statistical projects.
·
Finally,
map-related problems emerged in the operations stage, which were caused by
inefficient updating of the maps and inadequate training of the enumerators.
These difficulties, in turn, originated from the above-mentioned problems.
29.
Since
the completion of the 1996 census, the Statistical Centre of Iran has been
seeking strategies to overcome the problems encountered during the census, the
most noteworthy of which is providing the preliminaries for development of a GIS.
The project started six years ago and the progress is satisfactory with regard
to the financial and technical potentials. One step taken is digitization of
all rural maps in scales of 1:50,000 and 1:100,000.
30.
The
mapping staff in the SCI are working hard to resolve all the problems they
experienced in the last census and to get ready for the next one.
31.
In
the 1996 population and housing census, a post-enumeration survey was conducted.
It aimed at estimating underenumeration of population in terms of different
factors, such as lack of complete coverage of places and household members. The
results of the post-enumeration survey were not used for adjustment of the
census results. However, analysis of the PES results helped to diagnose the
factors contributing to the undercount, so as to prevent them from affecting
the 2006 census. It should be noted that the following changes were made in the
1996 census compared with the previous ones:
·
Decrease
in the number of lines of the household questionnaire;
·
Increase
in the number of pages of the questionnaire (from one to two);
·
Inclusion
of the identity card number and serial number in the items of the census
questionnaire; and
·
Household
lists existing in the post office of the Islamic Republic of Iran were used in
most urban areas.
32.
Some
of these changes increased the undercount of the population. Moreover, the
undercount in urban and rural areas as well as in different areas varied. It
seems that the analysis of the PES results may decrease the undercount of the
population in the 2006 census. We hope that the post-enumeration survey plan is
designed so that the results of the census could be adjusted.
