C A R I B B E A N                 C O M M U N I T Y               S E C R E T A R I A T




CENSUS COORDINATING COMMITTEE                                  RCCC/2005/13/2


Kingston, Jamaica

24-25 October 2005                                                                        18 October 2005




Please see attached, Paper entitled Report on the CARICOM Population and Housing Census Symposium, Nassau, The Bahamas, 16-17 September 2005.




















CHAIRPERSON                                                            DATE OF VENUE


Ms. Valerie Nam                                                 16-17 September 2005

Director                                                                        Nassau, The Bahamas

Census Demographic and

Social Statistics

Statistical Institute of Jamaica




The CARICOM Population and Housing Census (PHC) Symposium was held on 16 to 17 September 2005 at the Windham Nassau Resort Hotel and Casino, Nassau, The Bahamas.


1.                   OPENING CEREMONY


Following the singing of the National Anthem and an opening prayer, the Director of The Bahamas Department of Statistics, Mr. Charles Stuart warmly welcomed Delegates to the Forum and to The Bahamas.


Mr. Sylvan Roberts, Deputy Programme Manager, Statistics, Caribbean Community (CARICOM) Secretariat made some opening remarks.  He extended greetings on behalf of the CARICOM Secretariat and conveyed to the Government and people of The Bahamas their heartfelt appreciation on the occasion of The Bahamas hosting of this historic meeting. Mr. Roberts informed the gathering that the meeting was historic since it is the first of its kind but also since this is the first time in the history of census taking that the preparations and planning for this massive exercise are beginning so early. This is crucial for the ultimate success of the project. He then reminded the meeting of the many important uses of census data, stating that the PHC provides a 'goldmine of information that can never be over-exhausted'. He used the opportunity to call on the international partners in development to offer their full support in order to maximise the success of the PHC. 


In his remarks declaring the forum open, the Parliamentary Secretary in the Ministry of Finance, Mr. Michael Halkitis also welcomed the participants and extended greetings on behalf of the government and people of The
Bahamas.  The honourable gentleman reiterated the growing importance of accurate and timely statistics of a broad scope that can come from a census. He congratulated CARICOM for such an early start in the planning process for the 2010 round of PHCs.  He pledged his full support to the process.


The Opening Ceremony also featured choral renditions by "Friends 4 Life", a very popular local entertainment group.


Mrs. Kelsie Dorsett moved the Vote of Thanks.


2.                  ELECTION OF CHAIRMAN


Mrs. Valerie Nam, Director of Censuses of the Statistical Institute of Jamaica was elected as Chairperson of the Symposium.




The Agenda as adopted by the Meeting is at Annex II to this Report.


4.                   PROCEDURAL MATTERS


The meeting determined its hours of work.






The first two presentations were delivered by the Secretariat and UNSD representatives, respectively. The Secretariat representative's presentation introduced the background, objectives and expected outputs of the Meeting. The objectives of the Census Symposium were:


(i)                 to provide a forum to discuss the experiences of the 2000 Round of Population and Housing Censuses, with regard to problems, challenges and best practices encountered;


(ii)               The forum also usefully discussed post-enumeration activities including Processing of the Data and Analysis and Dissemination of the results;


(iii)             It is expected that the Symposium will also lead to the early addressing of issues with respect to the planning for the forthcoming Census decade 2010 by Member States, as well as the institution of effective regional support and a continuous programme of Population related research and analysis;

(iv)              The Symposium also helped to facilitate the identification of elements of a strategy that is cost-effective and that can enable the building of capacity in the CARICOM Region with respect to the management and conduct of Censuses and in the analysis and wider utilisation of the Census results. 


Among the outputs, the Meeting was informed that a framework for a regional strategy would be formulated.  The Meeting was also informed that the model used in the 1990 Round of Population and Housing Censuses and which was partially adopted in the 2000 Round would again be adapted to the current circumstances.  The RCCC will continue to coordinate activities at the regional level via a Project Implementation Unit that will be established at the CARICOM Secretariat.


The UNSD's presentation focused on the status of the planning for the 2010 Global Census Programme.  The UNSD representative reported that an Expert Group had been formed and met during August 2005.  The immediate Terms of Reference of this group is to conduct a thorough review of the United Nations Principles and Recommendations for Population and Housing Censuses.  This is one of the key goals of the Global Census Programme for the 2010 Round of Population and Housing Censuses.  The UNSD representative informed the Meeting that other key goals include ensuring that all countries conduct at least one Population and Housing Census during this round.  Thirdly, UNSD will make available a platform for the exchange of ideas and experiences, technical assistance, data dissemination and information.




Noted the plans for the regional approach to the 2010 Round of Population and Housing Censuses as presented in the paper by the CARICOM        representative;


Recommended stronger cooperation between countries of the region and between sub-regions;

Also recommended that Member States adopt practices and concepts recommended in the United Nations handbook on Principles and Recommendations for Population and Housing Censuses.




6.1        Overall Management of the Census Process: Planning,

Financing, Cost Considerations and Sourcing of Funds,

            Administrative and Legislative Issues and use of Contractors


Agenda Item 6 was then addressed by the Meeting. There were two presentations on Management and Planning by Montserrat and Jamaica.  The Montserrat representative informed that even though prior to their Population and Housing Census there were several activities aimed at monitoring the total population, mainly as a consequence of the volcanic eruptions, there was nevertheless significant support for the Population and Housing Census in 2001.  The census takers went through the usual steps in census taking, training the field staff and launching a very thorough public relations campaign.  Despite these efforts, the results were delayed causing anxiety among the public in view of the fact that they had offered the census takers a considerable amount of information.


 In her presentation, the Jamaica representative reported that management changes at the Institute lead to considerable challenges.  As a consequence, their Population and Housing Census were postponed from April 2001 to September 2001. This led to the need to outsource some of the necessary activities.  Several relevant committees were established to carry the process forward.  Their Population and Housing Census cost USD10M.  Two important considerations with respect to the outsourcing of activities were Capacity Development and Confidentiality.  In outsourcing the Scanning phase of their Population and Housing Census, it was ensured that the Institute staff work alongside the consultants in order to enhance capacity development. Also, in the outsourcing contract, a clause related to confidentiality was included. 


In the ensuing discussion, the Meeting noted the continuing high cost of mounting a Population and Housing Census and that the management of this massive exercise can be burdensome;


It was also noted that the timing of the census should avoid national elections or any other such activity that could negatively affect the smooth running of the fieldwork or the high quality of response to the questionnaire;


The discussion further noted the importance of a Regional Census Coordinator and Office to coordinate the Population and Housing Censuses successfully, as well as the early development of strategies to support census management;




Recommended that a regional approach to the conduct of the 2010 Round of Population and Housing Censuses be adopted;


Also recommended and endorsed by several participants, including the UNSD representative, that a core set of questions for the Region be developed that will form the nucleus of the regional questionnaire.  Member States may add further questions of national priority to their questionnaire bearing in mind that the final instrument should not be too burdensome for the respondent;


Urged Member States to institute early planning and should include a thorough review of the prevailing Statistics and Census Act.  This is particularly important in view of the fact that technology has significantly improved during the interim.


6.2              Effectiveness of Training for Census Field Enumeration - Training of

            Trainers, Supervisors and Enumerators - Implications for Data Quality


From the presentation on the 'Effectiveness of Training for Census Field Staff' the meeting noted that training at all levels is crucial for the conduct of a thorough and successful Population and Housing Census.  There was considerable discussion on the need to recruit and train intelligent and dedicated interviewers or enumerators. Such training should involve objective testing to assess capability of the enumerators.  The objective test can also help to 'weed out' the undesirables.


In the discussion, it was noted that comprehensive and concise training manuals should be designed to assist in the training, and to be retained for future reference and use by enumerators.


The Meeting:


Recommended that the RCO develops regional guidelines to assist the training at the National levels. 


Also recommended that training commences with the Training of Trainers.  This should be followed by the training of supervisors and enumerators;


Endorsed the suggestion that training be harmonised as far as possible;


Also endorsed the proposal that The CARICOM Secretariat develops a pool of expertise that the various Member States can draw on for training assistance, if so desired.


6.3       Census Pre-test - Usefulness and Experiences


A presentation was then made by The Bahamas on the Usefulness of a census pre-test.  The paper noted that there should be a comprehensive pre-test including tools, instruments, as well as processes.  Testing of certain aspects in one country and others in another may be explored.  The experiences gained can then be shared. This can help to foster the regional approach.  The objectives of the pre-test should be clearly defined.


The Meeting:


Recommended the pre-testing of all instruments and procedures adopted for use in the Census;

Also recommended the thorough pre-testing of the Census questionnaire, which is the principal data collection instrument


6.4             Review of Preparation of Census Documents and the Implementation of a System of Document Control and Flow


Barbados then presented their paper on 'A System of Document Control and Flow'. The paper emphasised the need for designing an efficient system in order to guide the management of the entire census. The flow of such a system can also reduce bottlenecks.


In the discussion which followed these two presentations it was noted that the census questionnaire should be one of the main instruments to test.  It was further noted that a range of tests including focus groups should be undertaken. This will help to refine the questionnaire at each level.


The discussion also noted the suggestion that 'Data Processing' should form part of the testing process. 


Finally, note was taken of the importance of reconciling the number of forms, especially the number of questionnaires, distributed and the number that are eventually returned.  A computerised system is recommended.




Recommended that Member States develop a comprehensive system of document control and flow;


Also recommended that this system be thorough tested, amended as necessary and used in the Census.





6.5             Preparation of Census Maps


The Meeting then addressed the very important challenges posed by the construction of good Field Maps. 


6.5.1       Conventional Approach to Preparing Maps of Enumeration Districts. Problems and Challenges and the Use of Conventional Maps in Outputs.


Jamaica presented a paper on the conventional approach to preparing ED maps.  Among other things, the paper emphasised the need for maps to be prepared for all levels of management e.g. District Supervisors' maps, Field Supervisors' maps as well as Enumerators' maps.


The paper further suggests that Mapping exercise should be ongoing. The Paper also called for regional guidelines in the area of Mapping and noted the fact that maps can be very useful as a medium to disseminate data.  Hence, capabilities should be enhanced to interpret and read maps and this should form part of the regional guidelines.


6.5.2       Experiences with the Use of Geographic Information System Capabilities in the Preparation of Census Maps and Implications for Data Dissemination


The use of GIS capabilities in the preparation of Census maps and the implications for Data Dissemination was then discussed in a paper by the Director of Statistics of Saint Lucia.  In his paper, the Director informed the Meeting of Saint Lucia's brief history, dating back to 1995, when the OAS assisted with the funding of two statistical assistants to assist with the development of enhanced maps for the expressed purpose of improving the data collection activities to be undertaken during the course of the Agriculture Census of 1995.  It was then realised that the hand-drawn maps were inadequate to guide enumerators with well, articulated reference information to locate.  This project was crucial for the Saint Lucia Statistical Office to establish a mapping unit in-house, and eventually in 1996/97, this lead to the establishment of digital mapping within the context of GIS software.


The system has been enhanced recently in 2004 through collaboration with the Ministry of Lands and Surveys who were able to obtain satellite imagery from a successful flyover of Saint Lucia.  This is helping to facilitate sample surveys of the island and will greatly assist in updating the digitised maps in preparation for the census of 2010.  The Paper further informed that Census data records have also been attached to the coordinates of each building in the census dataset which allows the authorities to define virtually any area in Saint Lucia within the GIS and report census data on this area.  The Paper concludes by strongly recommending the use and application of GIS in the upcoming 2010 round of Population and Housing Censuses.


In the ensuing discussion, it was noted that the popularity of the use of GIS for mapping and for data dissemination is increasing within the Region. As a consequence, it was further noted that efforts should be made at the regional level to investigate the feasibility of these systems for the 2010 Round of Population and Housing Censuses.




Recommended that early attention be directed to the update and/or production of Census Enumeration District maps for all levels of supervision;


Also recommended that consideration be given to the adoption of GIS maps where feasible.


6.6       Process of Determining the Content of the Questionnaire and the         Coverage of Themes- Sampling Approach to Specific Topics


The Barbados representative then made a presentation on the processes to determine the content of the questionnaire.  The method of deciding on a set of criteria in advance to help in deciding which questions should be retained was presented and several participants endorsed this method especially since it promotes transparency.


In the discussion that followed the presentation, the following were noted:


(i)                 Several participants reiterated the need for a regional core of questions.  Such core questions will be extremely useful  to guide policy and other decisions related to the CSME;


(ii)               If countries so desire, they can include additional questions to address their   individual needs;


(iii)             Some countries, in accordance with the experiences of  Barbados and Jamaica may also want to employ the 'Short' and 'Long' forms whereby the former is administered to all individuals and the latter to a sample;


(iv)              It was suggested that attempts be made to curtail the number of questions on the questionnaire as a means of curbing respondent fatigue.  This is crucial to the overall success of the entire exercise.  Over the years, the population in many countries have complained about the lengthy questionnaires that many Population and Housing Censuses currently use;


(v)                The importance of adequate testing of the questionnaire was again repeated by participants.




Endorsed the suggestion that additional questions should only be included if they satisfy the criteria adopted;


Recommended keeping the questionnaire as short as possible.


6.7       Critical Issues with respect to Data Collection


The Meeting then addressed critical matters relating to the data collection during the 2000 Round of Population and Housing Censuses. 


6.7.1   Public Support for the Census


The Director of the Statistical Office of St. Kitts and Nevis, in her paper on 'Public Support for the Census' emphasised the fact that every activity in a census should be geared towards achieving the relevant outputs that have been agreed.  The paper further emphasised, that public awareness and support are the key elements in achieving these outputs.  Public support is best obtained via a planned publicity campaign.  The paper continued with the note that publicity campaign should involve the media significantly, and could include, among other activities, talk shows, panel discussion, jingles, press releases, as well as the circulation of pamphlets with census propaganda, to schools and the wider community.  It was also pointed out that the census can be powerfully advertised through the school medium.  School children can play a key role here.  The paper strongly recommended a planned and comprehensive publicity campaign, designed well in advance and involving all stakeholders.




Noted that for every successful census an effective Publicity Campaign is essential.


6.7.2       Challenges with Recruitment and Retention of Enumerators


The Guyana representative then made his presentation on the challenges encountered in recruiting and retaining enumerators for the Guyana Census conducted in 2002.  Before the floor was opened to discussion by the Chair, a further three short papers were presented.  These focused on Census Supervision, Quality Control Procedures and Use of IT in Data Collection.


A lively discussion followed these presentations in which several participants shared similar experiences with respect to recruiting and retaining enumerators.


Several suggestions to address this problem were put forward. 






 Noted that:


(i)                 Recruitment methods must be as transparent as possible, providing an adequate level of remuneration, as well as reducing the number of questions on the questionnaire;


(ii)               Despite these efforts, challenges will remain especially in cases where the data are not always tangibly and visibly used in the interest of the people providing the information;


6.7.3   Experience with Census Supervision


The Bahamas presented a paper detailing their experiences with the Census Supervision. Among other things, the paper emphasized the need for quality management at every phase of the exercise. Different tiers of supervision were used in the Bahamas and this ensured that detailed attention was given even at the micro levels.


From the discussion that followed, it was noted that depending on the local needs, there should be different levels of supervision to adequately manage all operations for a successful census.


6.7.4       Quality Control Procedures


From the Paper on 'Quality Control Procedures' presented by Grenada, the need for good planning, management and supervision at all phases of census operations was noted.





6.7.5       Use of Information Technology in Data Collection


The Bermuda Paper shared their experiences from a sample survey using hand held PDAs to collect the data.  This Paper highlighted the challenges that they had.  Some of these included Data Security, the need for personnel, inclusive of enumerators, to be IT literate and the potential for electrical failure of the PDA. However, some benefits from this approach include improved data quality, reduction of manual editing and coding, high speed delivery of raw data and reduced costs overall.  In terms of the use of this method for census data capture, the paper suggested that the questionnaire be shorter and less complex.




 Noted the good experiences of Bermuda with the use of hand held devices in data collection;


Further noted that this may be the possible way forward.


6.8       Experiences with Data Capture and Processing - Criteria Governing the

            Choice of Data Capture and Processing


6.8.1       Experiences with the Manual Approach to Data Capture Choice of

           Editing Procedure and Outcomes


Three papers on 'Data capture and Processing' were then presented and discussed. The representative from Trinidad and Tobago presented their experiences as one of the only Member States who used conventional data entry to capture their Census 2000 data. The authorities in this Member State based their decision to use manual data capture since at the time, they had limited knowledge of the emerging scanning technology. In addition, their data processing staff were not adequately equipped or trained in this technology at the time. Generous use of IMPS 3.1 was made in order to process the data. One of the obvious consequences of using the conventional data entry method to process a census was the delay it caused in getting a final acceptable database from which statistical tables can be generated.  Hence, the usual census report was delayed. 


6.8.1       Use of Scanning for Data Capture and Editing: Challenges and Lessons Learnt


The Anguilla representative then presented her paper focusing on their experiences with the Scanning Technology. She highly recommended the use of Scanning in census data capture because of the speed and accuracy in getting results. Jamaica also did a presentation on their experience with the scanning technology. They warned that questionnaire preparation is a vital part of the success of the scanning procedure.




Noted that the use of Scanning to capture census and survey data is becoming increasingly efficient;


Also noted however that there may still be a place for conventional data entry for this purpose;


Recommended that the use of Scanning Technology be thoroughly explored with the objective of its potential use in processing the 2010 Round of Censuses.


6.9    Post Census Enumeration and Evaluation of the Quality of the

        Census Data


A paper and discussion on the merits and demerits of the PES then followed, the paper being presented by Suriname. The paper reiterated the fact that the PES is only one method of doing some evaluation of the Census, but that there are several limitations with the method. One of these is to do with the high cost of doing a PES. Another is the overall value that is derived from doing a PES. The Turks and Caicos Islands also shared their experiences with their PES that was conducted many months after the Census. Although all standard procedures were followed in their PES, the fact that it was done so late meant that the value of it is highly questionable.


In the discussion that followed, it was noted that the PES is often justified as a means of assessing the quality of the census data, but its strengths remain more in terms of assessing coverage.


It was also noted that the limited benefits that may be accrued from a PES versus the very high cost involved in doing it well, should therefore be seriously considered before a decision is taken to do the PES.


It was further noted that other methods of evaluating the data, such as comparing them with other data from external sources and/or the use of demographic analysis should also be explored, and very good management practices and procedures should be employed at all phases to assure quality data. In addition to the above, it was suggested that the Field Supervisors conduct a one hundred per cent check of all questionnaires returned to him within the first few days of the fieldwork, in order to ensure that the problems are identified as early as possible and corrected.  Field supervisors should also conduct re-interviews to ensure quality. Very focused management of operations will lead to a high quality of the data collected. This can help to avert the need for an expensive PES.




Recognised the high cost that is involved in doing a PES and the limited value that can be accrued from it;


Recommended that serious consideration be given to doing a PES and that alternative forms of evaluation should be explored before deciding to conduct a PES.


6.10    Preparation and Dissemination of Tabulations and Experiences with    

            Census Data Analysis at the National Level


6.10.1    Experience with the Preparation of Tabulations


Dominica was to present a paper on this sub-item but unfortunately, Dominica was not able to attend the meeting.


6.10.2    Census Data Analysis


On the topic of Analysis and Dissemination of Data Belize made two presentations. In her paper on data analysis, the Belize representative explained that analysis varies depending on the type of report required. A preliminary report should be prepared as early as possible after the fieldwork is completed. This is a very good public relations gesture. After the census data processing a general but more detailed census report should be done and thereafter, more in-depth analyses focusing on specific topics should be undertaken at the national level.




Noted the different levels of analysis that were outlined in this paper;


 Recommended that capacity be developed in the area of tabulation and analysis using software packages. Analysis should be an essential part of the census process. The use of the most appropriate medium and/or format for dissemination, including the use of the Internet is strongly encouraged.


6.10.3   Dissemination of Census Results


In her paper on Data Dissemination, the Belize representative advised that to maximize the effectiveness of data dissemination, it is crucial that preliminary planning in terms of a data dissemination strategy be done. This should involve 'what' is to be disseminated, 'how' it should be done, 'who' will prepare the information, when the information will be disseminated and to 'whom' will the data be presented.




Noted the importance of disseminating the data collected to promote their use;


Recommended that partnership be strengthened between academic institutions and statistical offices in order to foster more analysis and further dissemination of data collected in a PHC.


7.                  Regional and International Support to Member States


7.1   A Review of the Regional Strategy in the 2000 Round


A brief presentation on the regional strategy used in the 2000 Round of Population and Housing Censuses was then made by the Secretariat's representative. In this paper, it was reported that the model used in 1990 was replicated in 2000 with some adjustments. This was because the 1990 round was largely successful. However, in the 2000 Round, funding constraints seriously affected the performance of the model this time around.




Noted the major elements of the model; and

Further noted that it may again be used for the 2010 Round.


7.2       Insights into International Support - Prospects for the 2010 Round of



The UNSD representative then made his presentations on prospects for the 2010 Round of Censuses. Among other things he emphasised the benefits that can be accrued from the regional approach to Census taking. Among these this cooperation enhances harmonisation of outputs, improvements in cost efficiency, fills gaps in resource availability and provides opportunities to develop capacity for further activities.




Noted the presentation by the UNSD.





The Suriname representative could not attend the Meeting.




This item was subsumed under Agenda Item 6.9.






The Chairperson then quickly summarised the main issues including the recommendations that came out of the discussion and presentations at each agenda item.






The Meeting noted the various conclusions and recommendations adopted at the various agenda items. The Secretariat representative also informed that the findings from the Census Symposium will form an integral part of the regional strategy that will be formulated to guide the implementation of the 2010 Round of Population and Housing Censuses in the Region.  A consultant will be hired shortly in order to formulate the strategy. In connection with this process, the consultant will be visiting selected Member States and will make electronic contact with all others.




The Secretariat representative then summarised the various steps planned to move the process forward.  Among other things she informed the Meeting of some of the detailed plans for the way forward.





Called on Member States to note the immediate steps that will be taken in the preparation for the 2010 Round of Censuses.


Requested the fullest cooperation and assistance from Member States in this endeavour, to ensure that we develop a thorough and comprehensive blueprint of the way forward for the 2010 Round of Censuses.




There being no other business, the Meeting ended with the usual exchange of courtesies.





Designed and Maintained by
Regional Statistics, Caribbean Community Secretariat, Headquarters (Turkeyen) P.O. Box 10827 Georgetown, Guyana
Tel.: (592) 222-0001-75 Fax: (592) 222-0098
2014 The Caribbean Community Secretariat All rights reserved.