Hi all. This is a long shot but I have students who have submitted duplicate answers to essay questions and the only way I can pick them up is manually. Any ideas on querying the database to detect similar responses would be welcomed! I have tried exporting to excel and using the fuzzy logic addin but it just hangs - I think because of the number of fields or perhaps because it is not the right solution for the problem.
Any ideas?
Thanks
Karen
Hi Karen,
For a quick check, if you have access to the database via adminer or phpMyAdmin, go look at the mdl_question_attempt_step_data. Click on the field labeled, 'value' so the answers are all sorted. Scroll through the table, 'duplicates' will be bunched together. This is actually kind of limited as you will only be able to see a few words at the start of each essay answer, and if there are a LOT of quizzes, the table will be rather large.
A better way if you have the Ad-hoc Database Queries plugin stealled, would be to set up an ad-hoc query for that table and others, so you can have it also automatically tell you the name of the students for each entry. Plus, once you make the ad-hoc query, it is always just a couple of clicks away, any time you want to run it. As part of the
query for the database, sort the essay answers ascending or descending
and it will then put 'duplicates' together making them fairly easy to spot. You can also set up the query to scan for a particular quiz to cut down on the amount of results.
Hi Karen,
Sorry, I just don't have any lesson/essay data similar to that to try and play around with. Sorting via adminer or phpMyAdmin is actually very simple...just click on a field name and you're done. The number of records shouldn't be a problem as I have done sorts like this with thousands of records. The problem then is that the display of individual "text" fields tends to be truncated to a limited number of characters so if you have a bunch of "extra" data before the actual text, you would not be able to easily see what you need to.
Using the Ad-hoc Database Queries plugin, if you already have it or can install it, is a little more difficult to use, but is probably the easiest way to get usable data. It also makes it possible to do some filtering so that you might be able to get rid of all the "extra" stuff.
