I think the random block will select questions randomly from the random block.
For example i have a random block of 10 questions and want to present 5 questions to student.in this case student will not see all the 10 questions they can see only 5 questions selected randomly.
My requirement is, student want to see all the 10 questions but they need to answer only 5 questions they can skip 5 questions.
total point for the test will be 100 each question carries 20 point. So if they attend 5 questions of their choice they will get 100 point
2 of 2 people found this helpful
I don't think that Learn is going to be able to support this use case. There certainly isn't a way to keep the students from only answering 5 of the 10 questions. We can explore the use of extra credit maybe. What would you want the result to be if they answered 7 questions and got three wrong? Would you give them credit for 4 correct answers?
When I taught face to face, I'd do this for the essay portion of the test--give maybe seven questions, and ask students to answer five.
To address the "what if they answer more" my policy was that I'd grade the first five answers written. If they wrote on a question, and then decided not to submit that after all, they were to cross it out and write "omit" or "don't grade" on the writing--now a delete key would do the same.
In Bb, there's not a direct way I can think of to do this. But I thought of a couple ideas:
1. Could the questions be paired? Have five pairs, and have students answer only one of each pair? Would make grading a bit easier than my second idea.
2. Have six questions. The first lists all 10 essay prompts, and is worth 0 points. The next five are slots to provide the essay answers to their selected questions. For each question the student should include a number or other designator to indicate which prompt is being answered (trust me--necessary--sometimes you're not sure what they're trying to say.) This is less of a mess than trying to have them answer five essays in a single huge essay question. Feedback would be easier to provide too.
Keeping any sort of statistics on these would be difficult if not impossible, since students are essentially taking "different tests." But it's not unsound design, as far as I'm concerned.