Challenges in Census Data Processing Using Contemporary Technologies: The Case of Malawi
Angela Msosa
Demography and Social Statistics, National Statistical Office, Zomba, Malawi

Over the years there has been advancement in data capture methods for population censuses with the aim of producing the results in a shorter time frame. More and more African countries are adopting contemporary technologies like scanning for data capture. Malawi captured the 2008 Population and housing census using Optical Mark Reading (OMR). In a manual capture system, several details are checked on the forms that are designed to be read by humans including manual checking and setting up of the geographical codes existing at the time of the census enumeration. These processes tend to disappear with the emerging of scanning which is done on forms designed to be read by machines. This results in lag time between the end of the scanning and the production of a clean data file from which results can be run which is contrary to expectations that scanning produces final data in a very short time. Malawi's experience with the scanning process shows that there is need for sufficient planning before embarking on scanning more especially when this is to be done for the first time with the census. The pre-capture manual checks implemented in the manual data capture system tend to re-appear within the scanning system for census data processing at a different stage which is after the data capture process.

Keywords: Data; Processing; Scanning

Biography: Angela Msosa is the head of Demography and Social Statistics Division of Malawi NSO, at a position of Chief Statistician. She has been with NSO for over 10 years. She has widely worked on data processing on major surveys and censuses, her recent data processing work being the head of 2008 Malawi Census data processing team before she became head of Demography and Social Statistics Division.