Sometimes online participants break off before the end of the questionnaire.
If this is close to the end, their data can still be useful, since the survey routing will dictate that they must have completed all questions for which they were eligible until that point.
Similarly, postal participants sometimes break off, or in other cases miss questions within the questionnaire.
This can result in significant quantities of missing data, since routing cannot be enforced within a postal questionnaire.
We can include cases which don’t have complete data but if all demographic or all activity data are missing then they need to be excluded.
Sometimes completely, or almost completely, blank postal questionnaires are returned, or questionnaires which have been written across but not filled in.
Survey responses were checked at several stages to ensure that only cases with useful data were included.
At the initial data collation stage, the scanning team removed obviously blank paper questionnaires. Following this, during data processing, rules were enforced for the paper and online surveys to ensure that participants had provided sufficient data to make their data useful.
For the online survey, this meant participants had to reach a certain point in the questionnaire for their data to count as valid.
Paper data was judged by the completion of specific key questions that were vital for survey weighting and analysis.