Combining Data Sets Containing Character Variables of Different Lengths

[FreelanceReinh]

Fully agree with @Astounding. I would remove all the permanently associated $w. formats and informats from character variables (in a data step or with PROC DATASETS). I have never found these formats helpful in datasets (let alone informats). Please note that $w. (with w=length of the character variable) is the default format for character variables anyway. So, you gain nothing, but incur the risk of truncation and unnecessary "differing attributes" results in PROC COMPARE, if you assign this format permanently.