Data management

If one compares closely the details of the questionnaire and the data extracted and stored in the data file one can see that a fair amount of work was needed to code the data in a form suitable for analysis. For example, milk consumption needed to be first transformed into litres before data entry.

It was decided to ignore ages of members of the household, as this information proved difficult to collect, and to just enter the total amount consumed. Sour milk was not drunk and hence this column was ignored. The total number of children in the household was also calculated from the numbers given per wife.

Files CS1Doc1 and CS1Doc2 describe the variables contained in CS1Data1 and CS1Data2, respectively. Documentation is an important component of any investigation. Investigators are often dilatory in documenting their data but, if this is not done carefully, then this limits the possibilities for further exploitation of the data by another researcher at a later date. The documentation files provided here describe the data but have omitted descriptions of the sources of the data - when they were collected and for what purpose. Such information will need to be included when the data are archived at the completion of a project.