Nigeria - National Core Welfare Indicators Survey, 2006, Third edition
Reference ID | NGA-CWIQ-2006.v1.2 |
Year | 2006 |
Country | Nigeria |
Producer(s) | National Bureau of Statistics - Federal Government of Nigeria |
Sponsor(s) | World Bank - WB - Funding Federal Government of Nigeria - FGN - Funding |
Metadata | Download DDI Download RDF |
Created on | Oct 18, 2010 |
Last modified | Dec 02, 2013 |
Page views | 369343 |
Downloads | 136728 |
Data Processing
Data Editing
During scanning, the scanner took an image of each page of the questionnaire through form processing software (Teleform), which subsequently evaluated the scanned images. Evaluated images that suggested possible errors in the questionnaire were verified and corrected by the data entry operator. Typical errors included unidentified pages that could not be evaluated; unrecognisable hand printed characters or bubbles, which were not completely shaded. The time required for image evaluation and subsequent verification depended on how well and legibly the questionnaire was filled in.
After all potential errors for an EA had been verified by the data entry operator; the data from the questionnaires was transferred to a shared folder in the desktop computer. The output of the scanner was then checked for consistency, omission, skips and other errors; the data was not transferred to the database until all such errors were corrected.
Other Processing
The data processing was organized to run concurrently with the fieldwork. That is, the programmers and their assistants participated in the training of the field staff and remained in the field for a number of days to further guide the enumerators on how to print the response and shade the bubbles according to the code of responses. Hence, the main data processing operation commenced two week after the commencement of the fieldwork. The CWIQ system consists of the following phase: Data entry, data validation and correction, and generation of survey results. Data entry consists of converting the information in the survey questions to a readable form for processing in the subsequent phases. This data processing system was adopted from the system developed by the World Bank in 1998 and 1999 and subsequently used for the pilot survey of Lagos State in July and August of 1999. Some modifications were however, made in the questionnaire after the survey in Benue. The modified questionnaire was then used for the CWIQ survey in Abia, Cross-River, Ekiti, Kebbi, Kogi, Yobe, Jigawa and Enugu in May to August 2002. The States covered in year 2003 were repeated for the year 2004. These include Abia, Cross-River, Gombe, Kebbi, Osun and Plateau States. Further modifications was done to the questionnaire in may 2005, there was total overhaul of some sections and the reference number was pre printed and at the same time reduced to 4 digit; while a new methodology which used hand printing recognition was adapted. The data processing office for the current CWIQ was set up at the corporate headquarter of the National Bureau of Statistics in Abuja. Two processing centres were used and, the staff consisted of twelve (12) programmers and thirty four (34) computer assistants.
Twelve portable optical scanners each connected to a laptop computer were used for scanning the questionnaire on EA basis. The image evaluation and data verification were done simultaneously with the scanning of the questionnaire. Eight desktop computers were connected on a local area network (LAN) for error correction and data validation, and later converted to the database. Questionnaires coming from the field weekly by EA underwent identification and name confirmation at the processing centre using a master list of sample enumeration areas. The number of questionnaires and households for an EA were then counted to confirm that all the expected households in the enumeration area had been interviewed.
The questionnaires were then prepared for scanning by sorting the households therein in ascending household identifiers; the cover pages of the questionnaires were checked for completeness; staples were then removed from the questionnaires and the questionnaires for an EA were then scanned togetheroutput of the scanner was then checked for consistency, omission, skips and other errors; the data was not transferred to the database until all such errors were corrected. as a batch.