How to Make Multiple Choice Tests Lead to Even Better Memory
(and my help for FREE)
Most of what you need to know about crafting a good multiple choice test is available here from Inner Drive.
Read it if you do not know how to choose the best alternatives to the correct answer in a multiple choice test.
Now, what will make students remember more from your MCQ?
Inner Drive cite a recent study I did not know about.
And it could be a game changer. It goes like this:
Students who have to answer a question before they see the range of possible answers remember more.
There is an easy way to test this:
Take two parallel classes, or two equal halves of a year group (don’t rely on bands, as these are not usually of equal ability). One side does the standard multiple choice test. Let’s say there are 25 questions.
The other side gets the same 25 questions, each with a text box. The student types in their answer. Then the student gets the same 25 questions and multiple choice that the other half received.
Do this with 8 homeworks. That will create a long enough interval for us to test the impact on long term memory. I’m assuming the hand in rate will be the same. It is even better if you do the tests in class instead.
(Don’t forget to repeat questions at intervals to take advantage of the Forgetting Curve).
Compare the scores of each group in the homework quizzes (or class based quizzes).
Now give both sides a final test in class – sampling 12-13 multiple choice questions from each quiz, so that there are a total of 100 questions.
Compare the results.
You’ll have a pretty definitive answer as to whether this works.
No, the research tells us it will work.
You will have a brilliant answer as to how much it works. I think it will be a gamechanger. But I could be wrong. Only gathering the data will tell you.
My Free Offer
If you are in a school and want to try this, but don’t have time to make the quizzes – contact me. I’ll either train you, or do it for you. I’ll do it for free for the first two schools to reply. That should give us a large enough sample to know how it works.