Multiple choice exams:

The majority of teachers and students — myself included before researching this article — are under the impression that they should stick with their first instincts during a multiple choice test and not change their answer.

Over 70 years of research from more than 30 scientific studies on multiple-choice tests have shown that this belief is flawed. Generally, it seems people who change their answers on a multiple-choice test after further deliberation make the right decision. Justin Kruger and colleagues in a 2005 study published in Attitudes and Social Cognition, call it the “first instinct fallacy,” and explain that it is the result of something called “counterfactual thinking.”
Counterfactual thinking can best be described by the phrase, “If only.” This type of thinking can be seen in nearly every area of ones’ life. If you’re in the checkout line at the grocery store and decide to switch lines, it seems almost inevitable that the line you move to will slow down, and the line you were in will start moving faster. You think to yourself, “If only I had not changed lines, I could be out of here that much sooner.” Feeling frustrated with yourself for not sticking it out in the original line, the event becomes a memorable instance of when changing your initial choice was a bad decision.

According to Kruger, when it comes to taking multiple choice tests, it’s the more memorable and regrettable thoughts associated with changing your answer that have probably led you to think that changing your answer is generally a bad idea. To demonstrate this empirically, Kruger and colleagues conducted a series of experiments to tease apart the different aspects of the first instinct fallacy.

One of their experiments had specially-trained judges identify eraser marks on the multiple choice answer sheets from over 1,500 exams, and randomly selected 51 of the exam writers to offer their intuitions about whether the answer changes benefitted them, hurt them or were neutral — neutral being a change from a wrong answer to another wrong answer. Consistent with Kruger’s predictions and results from past experiments, exam takers correctly changed their answers from wrong to right by a two to one margin. Also consistent with memorable instances of counterfactual thinking, the exam takers surveyed overestimated the number of times they incorrectly changed their answer from right to wrong.

Their other experiments used different scenarios, including a mock Who Wants to Be a Millionaire game to substantiate the idea that people are better at remembering times when they changed their answer from correct to incorrect. The distinctiveness of these memories is what makes people think that sticking with their first instinct is a better strategy when taking tests, despite empirical evidence to the contrary. Knowing this fact apparently doesn’t change peoples’ behaviour, and they will generally still side with the lifetime of anecdotal evidence they have accrued and be hesitant to change their answer on multiple-choice exams.

Randall Jamieson, a learning and memory researcher here at the University of Manitoba, offered me a word of caution in the generalization of these results. He said, “In the absence of knowledge, intuition still provides the best guess.” This doesn’t mean, though, that you should ignore everything that’s written above. It means that, if you truly have no idea whether the answer is A, B, C or D, go with your gut; but if on reflection, you remember something that might indicate one answer over another, statistically speaking, you’re better off to go with that one.

Hopefully the next time you’re writing a multiple-choice exam, you’ll remember that it may be a good thing to change the answer from your initial instinct to the choice that, on deliberation, makes more sense.