Family Resources Survey 2001-02
Validation, editing, conversion and imputation
In addition to unit non-response, where a household does not participate, a problem inherent in all large surveys is item non-response. This occurs when a household agrees to give an interview, but either does not know the answer to certain questions or refuses to answer them (see the information on response for further information). They are still classified as fully co-operating households because there is enough known data to be of good use to the analyst.
The fact that the FRS allows missing values in the data can create problems for users. It was therefore decided before the first full year's FRS data was released that missing values should be imputed where appropriate. The policy has been that for variables which are components of key derived variables, such as total household income and housing costs, and areas key to the work of the Department, such as benefit receipt, there should be no missing information in the final data.
In addition to imputation, prior to publication, FRS data must be put through several stages of validation and editing. This is to ensure that the final data presented to the public are as accurate as possible.
The stages in the validation, editing, conversion and imputation process are laid out below.
