A. Introduction

1. The main objective of the Secretariat of the Pacific Community (SPC) Demography/Population Programme is to develop professional, technical, research and management capacity of the Pacific Island people in the areas of demography and population and development. The SPC Demography/Population Programme has provided population and demographic assistance to its 22 member countries and territories since 1967. The Programme supports Pacific Island governments in the collection, analysis, interpretation, dissemination and utilization of population and demographic data employing the appropriate and current international practices on population censuses and sample survey operations and procedures.

2. Three main strategies are used to achieve the above objectives:

· Technical assistance and advice with the design, collection, processing and analysis of data from population censuses and surveys;

· Training in demographic methods and analysis; and

· Research and dissemination of population statistics and derived development indicators for information, education, research, policy and planning;

3. In addition, the Programme collaborates in research and development and exchanges professional and technical information, views and ideas with other regional and international organizations.

4. The populations of the countries in the Pacific region range from a low of 1,500 to a high of over 5 million people, with estimated annual growth rates ranging from –3.1 per cent (due to international migration) to over 2 per cent. A number of countries have just completed taking censuses, while others will be undertaking censuses later in 2001, 2002 and the following years. This paper summarizes the Pacific experiences and views on the themes of the symposium.

B. Strategies for involving stakeholders in census activities

5. Since census taking is the most difficult and resource-intensive statistical undertaking in all countries in the region, the statistical offices established working partnerships with many stakeholders. Stakeholders play a vital role and support major census activities leading to the production of reliable and timely data that are readily available for internal use and external dissemination. The stakeholders were invited to be members of the various planning and coordinating committees within the overall census structure. These working committees were responsible for such major census activities as questionnaire content and design, field operations and logistics strategies, census mapping, publicity, and census product development. The stakeholders include relevant government organizations, non-governmental organizations (NGOs) and the private-sector organizations involved in research, resource allocation, policy and planning, law and order, communication and advocacy in various areas, as well as private business organizations.

6. The role of the representatives of stakeholders in the census coordinating and monitoring committees was important to the success of the census from planning to dissemination. For example, media representatives play vital roles in census publicity as well as information dissemination. Some private-sector organizations promoted or advertised census information and census dates in some of their leading products. In Papua New Guinea (PNG), Arnotts Biscuit Company promoted census 2000 messages and the census dates over six months before the census. Similarly, PNG Postal Services issued 2000 census postage stamps (with census logo) for public use 12 months before the census enumeration.

7. Stakeholders are often invited to submit in writing to statistical offices the kind of information they need from the population census or the kind of census product they would need from the census. Questionnaire design often takes these requirements into consideration. Stakeholders meet often and are informed by census administrators about the technical requirements for census taking, particularly in relation to questionnaire contents and respondent burden. Some of the methods used to limit topics in the census questionnaire include such questions as the level of detail required and the feasibility of collecting such information from censuses as opposed to specific sample surveys; the average level of education of the respondents and the enumerators; and, of course, resource considerations. Will the information be of acceptable quality for use? After these and related questions are considered and analysed, then the topic may be included in the census questionnaire, thereby meeting the client’s requirements. If client requirements cannot be met, then they are often informed of the technical reasons.

C. Strategies for choosing among data-collection methods as sources of demographic and social statistics

8. Population censuses and surveys will be the only source from which population and other socio-economic statistics that are nationally representative will be produced until vital registration systems and other administrative recording systems can produce nationally representative statistics. Quality and coverage are important considerations in making these choices. For example, registration of births, deaths, marriages, migration and so forth are not properly administered due to many factors including accessibility to these services. Moreover, members of the population do not see the need to register these events. These difficulties result in “no registration” in some countries and to “under-registration” in others.

9. In Papua New Guinea, as in many Pacific Island countries, birth, death, and marriage registrations do not occur at the time of the event but only when a birth or death certificate is required for education services or clearance of estates. With society recognizing traditional marriages, there does not seem to be any need to register marriages in many countries in the region. In fact, many of these vital events occur in rural areas, where registering of births, deaths and marriages is unheard of.

10. The resources available to statistical offices and the nature of the information required determine the choice between censuses and surveys. Of course, if there is donor interest in the information and funding is provided, then sample surveys are conducted. Most often in the Pacific region, censuses and surveys are either fully funded by donors or costs are shared between the donors and the government.

11. In the current census round, Tuvalu, a small Pacific Island country with a population of a little over 10,000, is currently seeking external funding and other technical support to conduct its census in November 2001.

D. Adapting new technologies to census operations

12. Countries in the region began adapting the latest technologies in census operations in the last round of censuses as well as in the current round. These technologies were used in the areas of communication, census preparation, census monitoring, processing and product development and dissemination. For example, in the most recent censuses countries have utilized e-mail, fax and fixed and cellular phones to relay messages between headquarters and provincial centres or outer islands. Also telephone hotlines were dedicated for census updates and information, while telegraphic money transfer services were utilized to transfer large amounts of funds to provinces and outer islands. Some countries used Global Positioning System (GPS) in census mapping.

13. In Papua New Guinea, field officers were issued with cellular phones, two-way radios and notebook computers with modem for e-mail and Internet access. Many of these field officers were not adequately trained to use these services efficiently, but over a period of three to six months through regular use and some guidance from others, many were able to use these technologies. Similarly, GPS units were purchased for provincial officers for census mapping purposes. Since the provincial census staff members in PNG have never used these systems before, not much use was made of the GPS.

14. Although most countries in the region would like to adapt to these modern technologies, many of them do not have the financial and manpower capacity to purchase and operate these technologies, as well as to maintain them adequately. As many countries depend on donor support for census activities, funding consideration is paramount in deciding on census work programmes, including outsourcing of some census activities. Many countries believe that outsourcing will not benefit census officials in terms of skills transfer in that particular area of operation.

E. Maintaining census-related activities during the intercensal period

15. Statistical offices in many countries document census procedures and operations from design to product development during the entire census operation. These documents are kept in libraries in-house as well as in research and academic libraries. However, there are constraints in many areas. Most of this documentation is done through technical assistance provided by donors as well as subregional organizations, such as the Secretariat of the Pacific Community. Many countries in the region have invested a lot of time and training to enable their census officials to document important census operations and procedures for future reference as well as to maintain institutional memory.

16. A related problem is retaining trained personnel after the census operation. Many countries (particularly small island countries in the region) have to start from scratch (in terms of trained personnel) with each census.

All final census data sets as well as related documentation are available for public use. Many countries lack proper storage facilities and the transition to new technologies leads to the problem of accessing historical data sets. The technological changes as well as the lack of proper storage facilities have also resulted in many countries’ losing past census data and documents.

17. Many census structures, especially field structures, are created on a temporary basis with external funding support. After the census operation is over, many countries close these temporary structures, as they can not continue to maintain them with only internal funding support. And there is no easy solution to this problem, as most statistical offices depend largely on government funding support. Oftentimes, the statistical office’s share of the national budget is sufficient only to pay salaries of permanent staff. Permanent staff members are those in the area of census design and planning, field operations, processing, analysis and product development. In many countries, the same group of personnel is involved in all these census operations because there are simply not many qualified persons to recruit. If there are persons with the relevant qualifications available, there is no position available due to staff and cost ceilings set by the government. When these people leave the census organization for better employment conditions elsewhere, the consequences to the census organization are enormous.

F. Identifying and resolving problems of census mapping

18. Census mapping is the single most difficult exercise for census organizations in the Pacific region. This is because of the lack of skilled personnel as well as lack of funding support to adequately cover all areas of the countries. Because of this, many countries experience difficulties in mapping. This leads to almost all map-related problems introduced in the census.

19. In the current round of censuses, some countries have collaborated with their National Mapping organizations to improve the quality of census mapping. In Papua New Guinea, the National Mapping Bureau did not fully understand census mapping requirements and placed a higher priority on their functional activities rather than census mapping. This led to delays in map-dependent 2000 census activities, and there was always misunderstanding between the two organizations. In the end, the census was conducted with incomplete maps.

G. Post-enumeration surveys: are they worth it or not?

20. Many countries acknowledge that a post-enumeration survey (PES) is important to evaluate census content and coverage errors, particularly the former. However, except for two or three countries, many countries have not had the resources to conduct a PES in the past and certainly not in the current round. Even if resources are available, many small countries believe that it is not worth the cost to conduct a PES.

21. For those that conducted the PES, it seems that the results are not properly analysed because of lack of proper institutional arrangement. Papua New Guinea undertook a PES and reported only the underenumeration factor after the matching process because there were no resources available to analyse the PES data. The census figures are not officially adjusted for underenumeration in Papua New Guinea.

______________

* This document was reproduced without formal editing.

** Secretariat of the Pacific Community (SPC), New Caledonia. The views expressed in the paper are those of the author and do not imply the expression of any opinion on the part of the United Nations Secretariat.