Multiple-choice quizzes appear in corporate training, schools, and online courses. They are quick to create, simple to grade, and compatible with any device. Evidence indicates these quizzes frequently assess the wrong skills and create misleading confidence in both learners and instructors.
Recognition versus recall highlights the issue.
When a quiz asks, “Which of these is a condition that triggers the escalation path?” the correct answer is visible. Learners do not need to remember the information; they only need to identify it. This task is far simpler and does not reflect how well they will apply the knowledge later.
Recall, on the other hand, demands retrieving information from memory without prompts. A question like, “Name the three conditions that trigger the escalation path,” requires greater mental effort. Research since the mid-2000s, including studies by Roediger and Karpicke, demonstrates that this type of retrieval practice improves long-term retention more effectively than passive review. Yet most quizzes avoid it.
High scores often mislead.
A multiple-choice question with four options guarantees a 25% chance of guessing correctly. If three options are clearly incorrect, even learners with partial knowledge can achieve high marks. A team might average 88% on a knowledge check yet still struggle to apply the concepts in practice.
Learners also overestimate their understanding. Karpicke, Butler, and Roediger found that people prefer study methods that feel smooth, such as easy multiple-choice quizzes, over methods that actually work. This leads to confidence without competence.
Incorrect answers can also linger in memory. Roediger and Marsh identified a “negative suggestibility effect,” where learners later remember wrong options as facts. Corrective feedback helps, but it is rarely included in the format.
Related: Nationwide plan to boost student outcomes
Improving the format is possible.
A study by Little, Bjork, Bjork, and Angello showed that multiple-choice quizzes can match the effectiveness of short-answer tests if the wrong answers are plausible enough to require real retrieval. The quality of distractors matters most.
Instead of creating wrong answers to fill space, use real mistakes from learners. These can come from support tickets, help-desk logs, or pilot tests. A distractor based on an actual misconception forces deeper thinking, turning one question into several retrieval attempts.
Other adjustments are straightforward. Ask an open-ended question first, such as, “What do you remember about X?” before revealing the options. Even ungraded, this primes the brain. Spread questions over days rather than clustering them at the end of a module. Always provide feedback explaining why wrong answers are incorrect.
The most difficult change may be accepting lower scores. A quiz that everyone passes does not measure learning—it merely goes through the motions. A delayed test, even a brief one, reveals what truly stuck. The resulting drop in scores is not a failure but the first accurate measure of real knowledge.
Training programs often treat quizzes as a formality. They are inexpensive, scalable, and maintain high completion rates. If the goal is meaningful learning, however, the format must demand more. This could mean fewer questions, greater effort per question, and a willingness to see lower initial scores.
For learners, the shift is harder. A quiz that feels easy may be the most deceptive—it leaves them believing they have mastered material they cannot use. The real test is not whether they can select the right answer from a list. It is whether they can produce it when it counts.
New approaches to professional training show how small changes in assessment design can lead to better outcomes.
