Making and using Variable labels - particularly with wide datasets

[rdstern]

I mention 4 datasets we are using for illustration

a) The MICS data is imported from SPSS and comes with efficient names and also variable labels.
b) The survey data from Kenya also has relatively short and organised names and also has variable labels
c) The recent procurement data has long names and does not have labels
d) the recent CCRP data is the same.

There are two tasks I would like us to consider.
Where the data has very long names and there are no labels, (and possibly a lot of variables?) it would sometimes be sensible to be able to move the name into the variable label. Either at the same time, or separately, perhaps the name could be shortened?

Might we do this in a dialogue, or in the grid, or might a new dialogue include a grid? Currently we have right-click at the side and bottom of the column metadata, but nothing for the columns (attributes, or in the meta-data grid itself. If we do have a right-click, then it is probably in addition to a dialogue - that's what we usually do?

Note that pasting (adding) variable labels should be a simple thing to do. Shortening the names could become complicated. This will probab ly use rename_with from dplyr. That implies a function be written and I hope we can start by doing this in a script.

There are more tasks - of course. The variable labels are all very well, but we need to be able to use them. Our ggplots should have options to use variable labels instead of the names. This is usually for the axis labels, lev[gent title, etc. And also for tables. We already do this in the One Variable > Frequencies etc, which uses labelled data.