Well, it's a few weeks later and our crew has a done a great job of helping me convert my files using the method described.
However, there are few items they found during the conversion process that I thought I'd pass along . . . in case any of you find yourselves in the same situation.
When converting from WebCT to Moodle be aware of these translation issues:
- In some of my randomized Question Sets (called Random Blocks in Moodle), I used fractional points for scoring. That is, for "review items" from previous tests I typically give .5 points per correct response (rather than 1.0 points). However, one cannot do this in Moodle. All items must be scored in whole points.
- To solve this issue, Dave (our Moodle Master) suggests changing all point values to whole numbers. For example, my .5-pt items all become 1.0-pt items. Then, all the items that were previously 1.0 point each should be converted to 2 points . . . to keep the ratio of scores the same as the original test. Then, you can either likewise adjust your course's total points to account for this shift or you can count the test for half of the number of points scored.
- I have some items that are in the "Matching item" format that have only two possible matches. For example, I have a list of six functions and ask students to select either "steroid hormone" or "nonsteroid hormone" for each of the six items. However, in Moodle one must have a minimum of THREE matching items.
- So Dave's suggestion is to simply add "Ignore this selection" as the third choice for such items.
- I think for some of them, I'll modify that a bit so that it's a more viable distractor based on the content.
- And when I get some extra time, I may change the items in question so that there really are three possible answers.
- In WebCT, some "Multiple Response" items graded as partial credit for selecting each correct item sometimes assigned 33.3% to each of three correct answers. That meant that WebCT really only have .99 points for a 1-point item. But WebCT then rounded up so that it really didn't make much difference in the long run. However, when converting to Moodle, each 33.3% designation became 33.333... instead. So it's actually more precise that WebCT. FYI.
- Short answer items (such as "fill in the blank") were not always case sensitive in WebCT (unless you specified that in the scoring). In Moodle, such items are always scored with case sensitivity.
- So if you want to accept either Eustachian tube or eustachian tube, you have to program in both answers as possible correct responses.