Randomized questions pools are not enough. We need to have randomized variables within each question in the pool to be effective.
"We conclude that randomization with pools of four (or even three) problems, which also contain randomized parameters, is an effective mitigation for collaborative cheating."
Randomization of questions selected from question pools and question parameters can neutralize the advantages of collaborative cheating in asynchronous exams. This study finds that when students are given the same exam question, even when parameters are randomized, cheaters performed, on average, 13% higher. Increasing the size of the question pool to 4 and randomizing which questions are selected from within each pool resulted in a statistically indistinguishable difference between cheaters and noncheaters. These findings suggest that the combination of the increased size of question pools and the randomization of both question selection within pools and the randomization of parameters can reduce the advantages gained through collaborative cheating.
When exams are administered asynchronously, the risk for collaborative cheating is high because students who take the exam earlier can share exam information with other students who have not yet taken the exam.
Randomization techniques should consider the appropriate level of randomization for both question pools and question parameters.
Increased randomization of question pools and question parameters makes it more difficult for cheaters to have access to exam information and harder for them to memorize the increased number of problems.