I recently started using the command line to correct MCQs. Here’s my setup.
The only prerequisites are the
nl unix utilities, along with a digitised answer key.
¶ Basic Usage
Let us assume two files:
a b a c d a a c b d
a b a c d d c c a d
You can easily diff them:
diff answer_sheet answer_key
6,8d5 < d < c < c 9a7,9 > a > c > b
However, this output is not very readable.
¶ Improving The Usability
nl to add line numbers to the files, along with some flags for
diff, we can get a prettier output (NOTE: This uses file descriptors and thus requires bash):
diff <(nl answer_sheet) <(nl answer_key) -y --suppress-common-lines
6 d | 6 a 7 c | 7 a 9 a | 9 b
As you can see, this output is much easier to understand, and only the mistakes.