Don’t Throw the Baby Out With the Multiple-Choice Bath Water

Multiple-choice questions are often maligned by instructional designers. I’ve heard the following at conferences.

• “Multiple-choice questions don’t measure anything.”
• “It’s too easy to guess them.”
• “Real life doesn’t have multiple-choice questions, so why should our course?”

So are they truly that evil? No, they aren’t if they are used properly. The problem is that it is quick and easy to write and program bad multiple-choice questions.  An unskilled instructional designer can look for simple facts in the content and ask simple questions about it.  These types of questions usually test memorization of irrelevant details rather than understanding and applications of the key concepts needed to perform the job.

When I first started in e-learning in 2001, I reviewed a course about what the Internet is. There was a little bit of history at the beginning, and then the course went on to cover some important concepts, such as the fact that the Internet is not a single place but rather an interconnected network. What was in the quiz? A multiple-choice question about when the Internet was invented. Unless you are on Jeopardy (as Rod was), when are you going to need to know the answer to that question? You aren’t. A better question would center around the characteristics that affect its use.

It is also very easy to “give away” the answer.  An inexperienced instructional designer focuses on the right answer but then doesn’t give much thought to the distractors—the incorrect answers. Oftentimes, they are so obviously wrong that it is extremely easy to guess.

Whenever I review off-the-shelf content, I always take the quiz first. I am amazed at how often I can pass a test on a subject I know nothing about. If I can pass a test on blood-borne pathogens without taking the course, then that’s a bad test.

These factors give multiple-choice questions a bad name, which is too bad, because they are a legitimate instructional tool. And like any tool, they should only be used for certain jobs. I wouldn’t try to peel a carrot with a blender or crack an egg with a potato masher.

So when do you use a multiple-choice question? When the real-life situation is also multiple-choice. For example, we just completed a course for the American Red Cross on how to work in a disaster shelter.  For a shelter worker, there are many situations where they have to make one of three choices: “Yes, I can do this.”  “No, I can’t do this.” “I need to check with my supervisor.”  So in situations like this, a performance-based multiple-choice question is a good fit.

If you focus on the behavior that needs to take place on the job and find that the student would have very specific options, then don’t be afraid to use multiple-choice questions to simulate that situation.

If you’d like to learn a little bit more about the design approach in the Red Cross course, check out the eLearning Guild’s Best of DemoFest webinar . Our course is featured about two-thirds of the way in.