Rwanda - Rwanda Population and Housing Census 2012

Start	End	Cycle
2012-08-16	2012-08-30	N/A

Face-to-face [f2f]
The Pilot Census in July / August 2011 took place in 75 EAs randomly selected from across Rwanda was a major test of the enumeration phase 'on-the-ground' in terms of the recruitment and training of enumerators, the listing of an EA, as well as the use of the questionnaire. In other words it was concentrating on ensuring measurement errors would be minimized in the actual census collection. After further consultation across the team, testing of ideas for fieldwork management was added to consider how to ensure response error would be controlled.
To enable the ultimate training of more than 17,000 enumerators and their team-leaders, the hierarchical structure of census management was used to cascade the standardised training down from the National and Provincial level to the local level. Core Master Trainers from NISR developed the training materials, including the CD of enumerator dialogues, trained those involved at the Province, District and Zone as Master Trainers. This was a two week residential training programme that included at least two fieldwork tests to assess and aid learning. These Master Trainers then cascaded the training down to the Sector level by replicating the two week residential training programme at training centres across Rwanda. Finally, the Sector level Trainers trained the team-leaders and enumerators using the same two week residential model, with the fieldwork tests, at 67 training centres. As training was cascaded down, the quality was ensured by the use of standard materials including the CD as well as by direct quality assurance at the training by the eight Core Master Trainers and the more numerous Master Trainers.
Careful planning by NISR ensured that everything was in-place to facilitate the actual enumeration of the population during the enumeration phase following census night, 15th August 2012.The enumeration of the population in private and institutional households was conducted across the country from 16th to 30th August 2012, just after the census reference night
The interviews were conducted in Kinyarwanda and English languages

A first draft questionnaire prepared by the National Institute of Statistics of Rwanda (NISR) was submitted to the Census Technical Committee for amendment before its approval by the National Census Commission. This version of the questionnaire and related manuals were tested during a Pilot Census in 2011. Lessons learnt allowed the NISR to finalize the questionnaire, containing 77 variables, and modify the manuals accordingly. The latest version of the questionnaire was subject to the approval of the National Census Commission prior to its administration to households. The questionnaires used to collect data were of two types: a private household questionnaire and an institutional questionnaire.

Name	Abbreviation	Affiliation
National Institute of Statistics of Rwanda	NISR	Ministry of Finance and Economic Planning

As initially planned, the enumeration of the population in private and institutional households was conducted across the country from 16th to 30th August 2012, just after the census reference night. Although data collection activities were carried out by well trained enumerators, quality assurance of the field work entailed close supervision by line managers at different levels:
- Team Supervisors, covering an average of five enumeration areas each
- Sector Controllers;
- Zonal Supervisors, covering between 2 and 5 administrative Sectors;
- District Coordinators;
- Province Coordinators;
- National Coordinators.
To quality control data collection phase, team-leaders had daily meetings with their team of three to five enumerators. This allowed reviewing of the last day's enumeration, planning for the next day, and as the enumeration phase progressed delivery of additional batches of forms if needed. Team-leaders collected completed forms each day and checked them for completeness, as well as re-visiting some households to check the accuracy, helping to control measurement error. The process of enumeration also allowed for some fine-tuning of the listing. This is a particular issue in some rural areas where at listing stage it can be difficult to understand exactly how many households exist in a compound, while this becomes clear when the compound is enumerated and can therefore result in either the addition or removal of households. A key component was then the daily feedback of numbers enumerated in each EA by the team-leaders using SMS, so that progress with respect to response error as measured by the listing could be monitored at all levels.
In accordance with the instructions contained in the census manual, each manager ensured smooth daily progress of the operations within his/her area of supervision. Enumerators were accountable for the work done on a daily basis to their team leaders who carried out the verification of completed questionnaires and brought appropriate solutions to the problems
encountered.