To properly document variables.
Each variable needs to be checked on the presence of variable labels, value labels and missing value codes.
Proper documentation of the variables consisting of:
Missing value code(s)
Executing researcher: To check whether each variable is documented including variable labels, value labels and missing value codes.
Project leaders: To advice the executing researcher about this topic.
Research assistant: N.a.
For the raw system files that have been obtained as described in the guideline File Maintenance, there needs to be an evaluation of the extent to which these files have been documented regarding the presence of variable labels, value labels and missing value codes. This needs to be undertaken and documented for each variable. Proper documentation ensures that co-researchers can also use the data and that the study can be properly archived at a later stage.
Problems can arise, in particular, when transforming from databases where label information is not available after exporting data to SPSS, for example MS-Access and SQL server databases. SPSS commands such as VARIABLE LABELS, VALUE LABELS and MISSING VALUES can be used to address this issue. If a proper codebook is already available for each registration form/questionnaire, then the introduction of variable and value labels in SPSS may be omitted. Missing value definitions however should, where applicable to a variable, always be included in the SPSS system file. The right labels and missing value codes can be linked to the variables using the available codebook.
Often data imported from (database) programs like Castor, NetQuestionnaires or Blaise already contain missing value definitions and also variable and value labels. This can be set relatively simply in the design of the input screen.
Have all variables been defined fully and clearly?
V3.0: 2 December 2016: Text updated
V2.0: 28 May 2015: Revision format
V1.1: 1 Jan 2010: English translation.
V1.0: 13 Apr 2004.