I am working on my first Hot Potatoes quiz, using the JQuiz format, and Multichoice/Multi-select option. I set up a simple quiz, and in testing I get a completely wrong response, see the screenshot. Also, is there no answer weighting available with the multi-select? It appears to be "no".
Thanks,
Chip
Another issue - I have set up a quiz with 10 questions, and it works in moodle, but it does not provide any scores. I set the overall weight at 100, and gave each of the 10 questions a score of 10, then in testing I answer all 10 questions correctly and receive a score of 0. With feedback turned on it shows a score of 0 for every answer.
What is the fix for this?
--
Chip
Hi Chip,
I can see you are going full steam ahead on content creation - good for you!
To help us understand what settings you are using where, please could you recreate your HotPot activities on my testing server, so that we can all see what settings you are using.
thanks
Gordon
That's cool, thanks for making that available to us, it's much appreciated.
I have my quiz on your server and all the settings match what is on mine.
Just a note - I have also set up two more quizzes, one is flashcards so no scoring there, the other is matching, and it DOES score correctly.
--
Chip
Hi Chip,
thanks for uploading your JQuiz file.
It seems that the individual "% correct" values that are available for each option in a question within JQuiz are not working as you expect.
I suggest you leave them at 100%, because that means "this is a correct answer". Anything less means "this a partially correct answer". Of course, 0% means "this answer is wrong".
I think if you set the question weighting for each question to 10, then you may get the behavior you seek, namely each question appear to be marked out of 10.
However, I think most believe leave both the "% correct" setting and the question weighting at simply 100%.
If you would lke to upload your JQuiz with multi-choice answers which you felt gave you incorrect feedback, we could have a look at that together too.
regards
Gordon
Hi, I have uploaded the multi-choice/multi-answer quiz, and have set no options, they are all the defaults. It is very strange in the way it is scoring - choose 1 answer (out of the 10 possible choices) and it gives you a completely wrong, irrelevant reply.
For example: the 10 possible choices are just 10 letters, the 5 vowels plus 5 consonants. In testing, I chose
1 consonant and pressed the Check button, and the reply box shows 4/10 Correct.
1 vowel and pressed the Check button, and the reply box shows 6/10.
4 vowels and one consonant the reply box showed 8/10 Correct.
Thanks for taking a look,
Chip
Chip
Considering your first example, "1 consonant", I wonder if the attached screnshot will help you understand why the score for that is 4 / 10.
For each of the 10 items you must ask yourself, should this item be checked or unchecked? If it is checked and it is supposed to be checked then it is judged to be correct. Similarly, If it is unchecked and it is supposed to be unchecked then it is judged to be correct.
On the other hand, If it is checked and it is supposed to be unchecked then it is judged to be wrong, and also If it is unchecked and it is supposed to be checked then it is judged to be wrong.
Hence, the score is shown as 4 / 10, which is completely the right score, and entirely relevant - according to my understanding of how the software works
Gordon
Hi Chip,
Can I ask what your pedagogical objective is with this exercise?
As far as I can see, this type of exercise functions as a test, not a learning exercise. Not only is there no individual feedback for each answer, the student can't even know which answers are right and which wrong. This can be very frustrating for students who are weak on the subject concerned if they are trying to understand it.
Like you, it took me a while to suss out how this question type works. This what it says in the help file: "If the answer is not completely correct, the student will see a readout of the number of correct choices, and one piece of feedback; this would be the feedback from the first item in the list which was either selected when it shouldn't be selected, or not selected when it should be selected" but I think Gordon's explanation above is clearer.
If it's hard for us teachers to understand what the scoring means, it's even more difficult for students to do so.
That's why I've never actually used this exercise type. I mainly create learning exercises and it's therefore essential that students know whether their response is right or wrong. Being able to provide feedback for each specific question is useful too. Even as a test, I prefer a straightforward multiple-choice format because I can more easily see which questions are haven't been understood by individuals or groups. (In fact, for testing I preferred Quiz: I used to write my questions in Hot Potatoes and then important them into Quiz where it was easier to analyse the results.)
I'd be interested to see examples of this question type successfully used for either learning or testing. I may be missing something.
Cheers,
Glenys
Thanks for the info and responses. As for the scoring, okay, so it works as designed, but it is completely non-intuitive to the student. If the student chooses 4 of 5 correctly and 1 incorrect then they should see a score of 80%, right or not? I'm no mathematician, so I may have that wrong, but it seems to me the scoring is not "user friendly".
Glenys, the example quiz is just a test of the scoring function to see if it would work as I was hoping it would. It's not meant for anything more than that.
I would hope to be able to make a quiz that asks the student to choose which letters of the given choices are vowels, or maybe which letters have the long "a" sound, etc. That would require a multi-choice/multi-answer question format.
There is now another, bigger issue to deal with though, so I will put that in another thread.
Thanks for the responses,
Chip
Chip, this has nothing to do with being a mathematician, it is pure logic. Gordon explains perfectly well the logic behind it.
You write "If the student chooses 4 of 5 correctly and 1 incorrect then they should see a score of 80%, right or not?"
Please, please do not just write a general idea, but give a concrete example of what you mean by "chooses 4 out of 5 correctly etc.". Just as Gordon did.
Joseph
PS.- However I agree with Glenys that the way this multianswer MCQ question works in HP is not very intuitive and perhaps should be avoided.
Hi Joseph,
Accepted, thank you, and yes, I agree, maybe it is best to avoid that quiz type, which is unfortunate, but, ok, I will work around that.
--
Chip
Chip,
there is another possibilty which you may prefer to the JQuiz multichoice. Might I suggest you try JMatch Sort.
This kind of quiz was devised and created by Agnes Simonet and Stan Bogdanov - great job guys!
If you re-did your JQuiz as a JMatch in the style of the attached jmatch.sort.jmt and displayed this via the HotPot or TaskChain modules in Moodle using the JMatch Sort output format, then students would have to sort the letters by dragging them from their initial position outside any box to drop each letter either in the "vowels" box or the "consonant" box. (see screenshot)
regards
Gordon
Hi Chip,
I totally agree with Gordon that JMatch Sort is a better solution to your Vowels and Consonants exercise. You'll find exactly how to do it on Stan's site here: Add-ons - and scroll down to JMatch Sort (89) v.1.4.3. It assumes you know how to install add-ons, see:
Here's a working exercise: Letter A - Practice Mode.
Cheers,
Glenys
Greets from a lurker,
First, thanks to Glenys for pointing to my site.
This is slightly off-topic but is related to the multiselect type of exercise.
I also avoid using it, but once in a while it turns out useful. I'll exemplify with a screenshot at the same time trying to clear the confusion around this question type.
In this question, what I want from students is to identify the two possible answers.
Our 'poor' logic is that we (most people - and we also teach students that way) confuse scoring with logic, as it was mentioned. You can see in the screenshot 3/3 for feedback. This is because the student has made three correct decisions - a) and b) are correct i.e. possible answers and therefore checked, answer c) is wrong and therefore unchecked. The total score on the question is 100% because the student has made three right decisions.
The screenshot also illustrates how I've used the multiselect. However, if one wants to stick to the popular logic of selecting the one and only when having two or more correct alternatives, he may use Multiple choice and word his question negatively. In my example, that may have been "Which of the three is NOT correct?" and be displayed as a multiple-choice question. BTW, students need a little training how to think about and interpret such scoring.
A more pedagogically meaningful exercise is to train students to predict what they may hear in a recording or read in a text within certain context. I use it in listening and reading comprehension activities to make students READ the questions more carefully before they hear the recording or read the text. This is a modification that's not listed in my Taxonomy and requires some hacking. It looks something like the second screenshot. It's not the best example, but still OK to teach students to expect and predict from context ...
Cheers
Stan
