Population and Housing Census 2002

Staffing and recruitment
The recruitment of external personnel was somewhat restricted due to time and financial constraints. To address this issue, the Census Office invited any staff from the public service who had past census experience to apply. Recruits were mainly from the public service and teachers. Overall, 15 supervisors and 60 enumerators were hired.

The area requiring the most staff resources was ‘Location’, which contained 103 housing blocks. Each block contained eight units or flats, with each of these enumerated as one household unless otherwise stated by the occupants. As well as these 103 blocks there were just under 20 ruins, bringing the total to over 600 habitable houses. The added problem of people speaking different languages housed randomly throughout Location meant recruiting enumerators who could speak various languages.

Mapping and listing operations
Mapping and listing were part of the preparations completed before the enumeration phase on 23 September. Two Census Office staff did preliminary household listings and mapping over a period of two weeks during early 2001, and then again during February 2002 due to a housing construction scheme that had been undertaken jointly by the Taiwan and Nauruan governments, and repatriation of housing formerly occupied by foreign workers. Mapping and listing operations proved very time-consuming in Location due to communication problems with the many different ethnic communities residing in the area. At the same time, the most recent maps (1992) were acquired from the Nauru Rehabilitation Corporation (NRC) and updated.

Mapping
The Office of Lands and Survey was the obvious first choice to provide maps. Unfortunately the size and scale of their maps made it impossible to create copies with existing copy machines on the island. This led to the second option of approaching NRC, who provided their most recent maps using some of the latest technology available. Throughout the census preparation and up until the day before enumeration, NRC provided the necessary expertise with census maps and production, incorporating rough locations of new houses.

Each district was divided into four or five enumeration areas (EAs), depending on the number of houses. Each enumerator was required to enumerate 10-14 houses. The recruitment of specific staff for Location meant further improvising because of the language barriers mentioned earlier, which required good and constant communication between each enumerator and his/her supervisor. Realising the challenges and time involved in maintaining such ongoing coordination and communication, it was decided to place a second supervisor in this area. The SPC demographer, Dr Chris McMurray, volunteered to assist in Location.

By enumeration day, all enumerators were provided with a map of their respective areas, with demarcations incorporating any changes they themselves had previously made. Each supervisor was provided with copies of the enumerator maps of the area they were supervising.

Household listing
Preliminary household listings were completed during the preparation stages. These listings were compared to the maps for credibility. Updating at a later time meant only minor changes were necessary during the first visit by the enumerators on 22 September. On each occasion, it took census staff around two weeks to complete these tasks. The actual process of allocating EAs within districts, and then listing the respective household heads, was important to prevent double-counting - of which, incidentally, no cases were reported.

Publicity
Publicity played an important role in the census operation. In order for any census to be successful, widespread publicity has to be achieved. The aim of every publicity campaign of this nature is twofold: to raise public awareness, and to educate. On an island as small as Nauru, such a campaign was organised with relative ease.

The Census Office decided to rely on three means of communication, which were also used for the 1992 census. The first was to publish a teachers' manual containing the most basic information about the census, its application and its importance to planners and governments. This manual was a modified version of the 1992 teachers' manual. The purpose was to incorporate census awareness into the school curriculum. The second means of communication was using radio and television to broadcast the message. The third was through the local medium known as the 'coconut wireless' or word of mouth, and depended entirely on the success of the former two and on census staff. Publicity was done during the training, with a crew from Nauru Television sent to cover two days of the training. On these two occasions selected participants of the census-training workshop were interviewed, and a prepared speech was given by the Acting Assistant Government Statistician describing the importance of the census and the benefits to the people in terms of future planning.

Unfortunately it was not possible to achieve the same scale of publicity enjoyed during the 1992 census. Although enumerators reported that none of the houses enumerated was ignorant that a census was being conducted, the fact remains that messages conveyed over the radio and television did not reach everybody as expected. It was found that households located in the northern part of the island did not receive radio or television transmission due to poor infrastructure. This situation contributed to some tension and hostilities during the enumeration phase, as well as disinterest and in some cases the temporary disappearance of entire listed households. Despite these obstacles, overall non-response was very low.

Training
Training of census supervisors and enumerators was conducted jointly with the assistance of the former SPC demographer, Dr McMurray. The training took approximately two weeks (including listing and training of additional enumerator assistants) to ensure all aspects were adequately covered. The sessions comprised three days of lectures, one half-day of field testing the questionnaire, five days of listing and two extra days of training enumerator assistants due to late staff withdrawals. Sundays were days off. Due to the length of the questionnaire, the group was pressed to cover everything adequately.

The same training manual was provided for each recruit and the questions were addressed one by one, followed by thorough discussions of the topics covered. With everyone exposed to the same training, supervisors were selected for their demonstrated initiative and their general understanding of the questionnaire, as well as for previous census experience.

Census
Census night was Monday 23 September 2002. The first stage of enumeration began the day before. This was known as the first visit, where enumerators visited each house within their allocated EA and listed every individual living within the house. The first visit also allowed enumerators to make changes to existing maps and household lists if required. Any changes to the maps were handed to NRC, who made the necessary alterations. The Census Office and the respective supervisor updated their own lists where necessary. During enumeration, each enumerator kept track of enumerated households using their maps and household listing, or field control sheets. Each supervisor was provided with a control sheet identical to the field control sheets, and maps that he/she was required to check for consistency. Once they completed work in their respective EA, the enumerators were required to hand in all forms to their supervisor, who subsequently checked for completeness and quality of the information provided, returning any unsatisfactory form(s) to the enumerator for re-enumeration. All these operations were coordinated by the Acting Government Statistician, who visited each supervisor during the field operations. Once all forms had been quality-checked by the respective supervisors, they were submitted to the Census Office together with field control sheets and a summarised control sheet. The control sheets were collected and entered into a computer, and a provisional count of the population was made.