Note:
...
The PDF should be used for entry and verification.
Parsers used to generate the xml required to import the questionnaire into Archivist: USoc exported XML is used as input for parse_understanding_society. Archivist tables are produced as the output and then used as inputs for archivist_insert.
...
XML is produced as the output.
...
- Ensure the most recent version of the PDFs should be used for entry and verification.Include the PDF is being used (check the PDF we have against the study website).
- There are two versions of the questionnaire in one (web mode and telephone mode. Do not add questions or sections based on telephone mode.
- The correct sequence and study wave is included in the qc/qi labels.
- If there is a missing response domain, check the previous waves to see if it's included and use that.
- Create specific numeric response domains where possible. Search for the question in the XML to see if there is a min and max. If no "min" and "max" is shown in xml XML file, then use "How many" . See example 1 below.
- The entire Proxy section should be entered inside a loop, as Proxy questions may be answered for more than one missing household member.
- If there is a missing response domain, check the previous waves to see if it's included and use that.
- (see Example 1).
Do not enter mixed-mode interviewer instructions (see Example 2).
- Do not enter soft or hard checks.
- Do not add questions or sections based on "Telephone mode".
Example 1
Question as presented in the pdf file with no range of of days specified:
- Note that question literals and code list categories containing { } brackets do not translate correctly into Archivist - please copy and paste the correct question literal or code list manually (issue under review).
Example 1 (us5)
XML file specifying the "min" and "max" range associated with the question:
Example 2 (us5)
