Questionnaire versions
As described in earlier sections, the data have been collected from two sources: an online questionnaire and a paper questionnaire.
The online questionnaire includes some built-in routing and checks within it, whereas the paper questionnaire relies on correct navigation by respondents and there is no constraint on the answers they can give.
In addition, the online data are available immediately in the office in their raw form; however the paper questionnaire data must be scanned as part of a separate process.
Tick box answers are captured by scanning while handwritten entries are manually entered. These data are delivered in weekly batches.
Despite the quality control checks conducted on the data, there is potential for errors to be introduced during the process of scanning and manual entry.
All this means that in the early stages of the data processing, the paper questionnaire data are handled separately from the online data and are also subject to a higher level of checking and editing than the online data.
Within the online questionnaire there are two routes through the questionnaire, each taken by half the online sample, with some questions only asked by group 1 and some by group 2.
This section describes the processes by which the data were cleaned and edited, merged together and duplicates removed.
These different questionnaire versions also have implications for how missing data are handled and described and how derived variables are created.
Month of data collection
The fieldwork section has described how the survey sample is divided into waves and issued once a month.
The fieldwork cycle for a single wave lasts about five to six weeks. The data analysis involves cutting off cases by month of response to the questionnaire.
For data analysis, there are 12 months, starting on the 16th of the month and finishing on the 15th of the following month (November 2023 to November 2024).
Within the data for one month, there may be cases issued over several different waves. For example, in month 3 (January 2023 - February 2024) data could come from surveys distributed during waves 97, 98 or 99 of the survey.
The date of completion of the questionnaire is important for the way in which data are handled and weighted. Online data come with an automatic date stamp and it is known exactly when the questionnaire was completed.
Where a valid date of completion was put on the paper questionnaire, consistent with the timing of that case being issued and the questionnaire being returned, the date given by the respondent is taken as the date of return.
In some cases, respondents give no date, give an incomplete date, or give a date which is impossible (e.g. after the date the questionnaire was received in the office or before the questionnaire was sent to the respondent).
In these cases, the date of receipt is assumed to be two to three days before the date the questionnaire was processed by the scanning agency.