6 August 2001
Symposium on Global Review of 2000 Round of
Population and Housing Censuses:
Mid-Decade Assessment and Future Prospects
Department of Economic and Social Affairs
United Nations Secretariat
New York, 7-10 August 2001
Statement from Hong Kong
Special Administrative Region of China *
Alvin W. K. Li **
1. It is established practice since 1961 for Hong Kong Special Administrative Region of China to conduct a population census every 10 years and a by-census in the middle of the intercensal period. Following this practice, Hong Kong Special Administrative Region of China has conducted its latest population census (i.e., the 2001 population census, also known as 01C), from 15 to 27 March 2001.
2. Population censuses and by-censuses are not the only sources of population data in Hong Kong Special Administrative Region of China. Apart from them, results of sample surveys on a smaller scale and statistical data compiled based on information from administrative systems are also the basic data for compiling various types of population figures. While data from different sources may serve different purposes, in combination they form a comprehensive system of population data.
3. Broadly speaking, population census and by-census data are benchmarking population data, while also serving as the prime sources of data for small areas and population subgroups. These benchmark data are taken together with statistical data from administrative systems (such as birth, death and passenger-movement records) and sample surveys (in particular, the continuous general household survey) to provide a population statistical database which serves a multitude of purposes.
4. Population statistics in Hong Kong Special Administrative Region of China basically adopt the “resident population” approach; accordingly, this approach was adopted in the 01C. Under the approach and given the residency and mobility of the population, the “resident population” comprises “usual residents” and “mobile residents”. Usual residents refer to two categories of people:
(a) Permanent residents who have stayed in Hong Kong Special Administrative Region of China for at least three months during the six months before or for at least three months during the six months after the reference time-point, regardless of whether they are in Hong Kong Special Administrative Region of China or not at the reference time-point; and
(b) Non-permanent residents who are in Hong Kong Special Administrative Region of China at the reference time-point (These are people given legal status for a limited period of stay.).
As for mobile residents, they are permanent residents who have stayed in Hong Kong Special Administrative Region of China for at least one month but less than three months during the six months before or for at least one month but less than three months during the six months after the reference time-point, regardless of whether they are present or not at the reference time-point. Under this approach, visitors are not included in the population of the Hong Kong Administrative Region of China.
5. In the 01C, six sevenths of the households were subject to a short-form enumeration to provide some basic information like age and sex. The remaining one seventh were subject to a long-form enumeration, in which, in addition to basic data, information was also collected on a broad range of socio-economic characteristics of household members, such as educational attainment, occupation, income and marital status.
6. The short forms were mailed to the householders a few days before the census period for self-completion. Enumerators then visited the households during the census period to do a simple check and collect the completed questionnaires. For the long-form enumeration, the “interviewer method” was used. Enumerators visited households to conduct face-to-face interviews to collect the required information.
7. About 22,000 temporary fieldworkers (TFWs) were employed for the 01C. They were all teachers and students. This arrangement was adopted having regard for the following considerations: stable supply of manpower, effective and continuous training and close teamwork among TFWs.
8. Extensive consultations were held with users in both the public and private sectors in drawing up the list of data topics. Two pre-test surveys were conducted to assess the technical feasibility of the proposed new questions.
9. The main considerations in deciding the topics for the 01C were: (a) usefulness of data; (b) willingness and ability of the respondents to answer; (c) ability of enumerators to understand the questions and communicate with respondents; (d) existence of another data source or data-collection channel; (e) processing efforts; (f) keeping the interviewing time within a reasonable length; and (g) recommendations of the United Nations and international practices.
10. A post-enumeration survey (PES) was conducted shortly after the 01C to collect information for evaluating 01C results. A random sample of 15,000 quarters were revisited by regular field staff of the Census and Statistics Department. The PES results are now being analysed. Apart from the PES, demographic analysis and comparison against independent sources are also adopted where appropriate to evaluate the 01C results.
11. Up-to-date and comprehensive digital maps for the whole territory of Hong Kong Special Administrative Region of China are already available. The 01C has made use of these digital maps, with suitable modifications and adaptations, to support a whole range of census activities, such as production of maps for field operations, allocation of assignments to enumerators, monitoring of fieldwork progress, and dissemination of results. It has been proven that the use of digital maps could improve both the efficiency and effectiveness of the census activities.
12. On data capturing, bar-code technology was adopted for the reference number of each questionnaire and Optical Mark Recognition (OMR) technology for the answers in the long and short forms. Key-from-image technology, supplemented with computer-aided coding, was adopted for answers requiring specialized coding, such as occupation, industry and place of work or study. The method adopted in the 01C proved to be a very efficient means of data capturing; some 2.65 million questionnaires were input in 51 working days with an accuracy rate of almost 100 per cent.
13. The information collected in population censuses and by-censuses is vital to government planning and policy formulation, particularly in such fields as education, housing, transport, medical and social services. Data on small areas form the basis of district development and service-planning programmes of the government, while comprehensive information on population subgroups, such as the elderly and single-parent families, facilitates the planning of welfare policy and social service programmes. Most important of all, results of the censuses and by-censuses are the benchmark data for the compilation of population estimates in the population data system.
14. The system for compiling population estimates is complex and requires data from a wide variety of sources. The results of population censuses and by-censuses form the cornerstone of the system. These benchmark data are taken together with statistical data generated from administrative systems (such as birth, death and passenger-movement records) and sample surveys (in particular the continuous general household survey) to form a population statistical database which serves a multitude of purposes, including the compilation of population estimates.
15. According to the schedule, the Census and Statistics Department will release at the end of August 2001 the provisional population estimates for mid-March 2001 (i.e., the census moment) and for the end of June 2001. Basic information collected in the 2001 population census will be used in the compilation of these estimates. However, since “resident population” refers to “usual residents” and “mobile residents”, and the compilation of statistics on “mobile residents” requires data generated from the passenger-movement record system for the six months after the reference period, the revised population estimates for mid-March will not be available until October (i.e., upon the availability of complete information on passenger movements up to mid-September).
16. Other detailed statistical data compiled on the basis of the 2001 population census will be released in phases beginning in the latter part of October 2001, starting with a set of basic tabulations. The revised population estimates for mid-March will be released in parallel with other data in October. As for the revised mid-year population estimates for 2001, they will be released in accordance with the usual release schedule, that is, in February 2002.
17. To cater for the wide range of needs of different users from government, academic institutions, business firms and the community, the data of 01C will be made available in a variety of forms, including press releases, publications, CD-ROMs, value-added CD-ROMs and sample data sets.
18. A series of reports presenting the results of 01C will be published beginning in October 2001. Given the growing popularity of electronic media to disseminate results and the increasing demand for electronic copies of tables, the number of 01C publications will be trimmed down. However, the contents of the publications will be enriched and new publications will be included to meet user needs better.
19. Two basic CD-ROM products at low cost are planned: one contains statistical tables on a broad range of socio-economic characteristics of the population and the other contains digital maps of Hong Kong Special Administrative Region of China for geographical presentation of statistical data contained in the former. Value-added CD-ROM products, which bundle the 01C results with the digital maps in an integrated manner and provide users with tailor-made utilities for retrieval and use of the data, will also be developed.
20. A user-friendly dynamic reporting system is being developed for the 01C. Users can make use of it to compile non-standard tables easily and efficiently.
21. The Internet is now a very important means of data dissemination. A set of core tabulations on the 01C will be put on the Internet where they will be available for users to download at a charge.
* This document was reproduced without formal editing.
** Census and Statistics Department, Hong Kong Special Administrative Region of China. 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
 A by-census differs from a full census in that it does not have a complete headcount of the population but rather enquires about the detailed characteristics of a large sample of the population.