You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Duplicate variable labels might arise in two ways. First, Survey Solutions' variable labels for multi-component variables might be the same for all components if the variable label reaches the 80-character limit before the component label can be added. For example, a multi-select question might have question text used as its label that is so long that the multi-select answer option component does not appear in the variable label. Second, data users might accidentally apply to the same variable label to more than one variable.
To identify these issues, one needs a means of listing variables with duplicate variable labels and, potentially, of asserting that no duplicate variable labels exist.
Implementation ideas
Somehow, one would need to create a data set of varname and varlabel. From there, it's a matter of removing missing labels and using duplicates, list to communicate where duplicate variable labels lie. This would need to occur with preserve - restore combo to revert to the data in memory before looking for duplicate variable labels.
The uselabel command does something similar, but only for value labels.
The describe, replace does what's needed, replacing the data in memory with the contents a data set of variable metadata.
The text was updated successfully, but these errors were encountered:
Problem
Duplicate variable labels might arise in two ways. First, Survey Solutions' variable labels for multi-component variables might be the same for all components if the variable label reaches the 80-character limit before the component label can be added. For example, a multi-select question might have question text used as its label that is so long that the multi-select answer option component does not appear in the variable label. Second, data users might accidentally apply to the same variable label to more than one variable.
To identify these issues, one needs a means of listing variables with duplicate variable labels and, potentially, of asserting that no duplicate variable labels exist.
Implementation ideas
Somehow, one would need to create a data set of
varname
andvarlabel
. From there, it's a matter of removing missing labels and usingduplicates, list
to communicate where duplicate variable labels lie. This would need to occur withpreserve
-restore
combo to revert to the data in memory before looking for duplicate variable labels.The
uselabel
command does something similar, but only for value labels.The
describe, replace
does what's needed, replacing the data in memory with the contents a data set of variable metadata.The text was updated successfully, but these errors were encountered: