Multiple choice questions have a bad reputation. They are seen as being simplistic, easy and aimed solely at creating grades rather than testing understanding. I disagree. Those things are all true, but multiple choice questions can be a very useful tool when they are done well.
I recently came across this site which gives advice for writing good multiple choice questions. The best thing is that these principles are easily understood, so students can be given the task of writing the questions. Furthermore, by writing good multiple choice questions about in-class topics, students are practicing that essential fieldwork technique, questionnaire writing.
The techniques for writing good questions can be split into the stem (the ‘question’ part) and response.
Writing good stems
A good stem should not require a multiple choice response. It should be answerable without seeing the potential responses. This means that rather than choosing the best fit, the student will instead identify their response and then select the most closely matching choice.
It’s possible to write good stems that have more than one ‘correct’ answer in the responses, but there is a best choice of answer. For example: Where is the Eiffel Tower? The responses could include: (a) In France (b) In Paris (c) South of London. Clearly, the best response is (b) Paris, because it is the most specific. A good stem will make the criteria clear. The question should be: What is the most specific location of the Eiffel Tower from the list provided? Alternatively, students could be asked to select all the correct answers that apply.
Some students will want to reference a resource, such as a graph or map, as part of their question. If so, they should make sure that the resource is directly relevant. It is useless to give a map of London and ask ‘If a map of Cape Town was at a similar scale to this one, what would be the distance in km of 3cm?’. Any graphics should also be extremely clear – and fully referenced, of course!
Finally, questions should be based on objectively verifiable information, and not be opinion based. Naturally, the exception is when you are canvassing student opinions – but in this case, there is no wrong answer!
Writing good responses
There are two types of response: the answers, and the distractors. The answer is the correct response, while the distractor is the incorrect response.
Good answers are specific, and roughly the same level of detail and length as distractors. They should be the only plausible answer to the question, unless the question specifically states that more than one answer is to be chosen. A bad set of answers would be ‘What is your favourite colour?’ (a) Red (b) Green (c) Yellow, because it doesn’t allow the possibility of the correct answer being not on the list. However, it’s best to avoid the answer being ‘none’ or ‘all’ of the options, because again this doesn’t challenge the student – they can likely work it out.
It’s not quite needless to say that the answer should always be correct. It should never be a ‘close enough’ approximation to the real answer. If a margin of error is factored in to the answer, this should be made clear in the question.
Distractors should always be potential answers. Silly responses such as “‘Where is the Eiffel Tower’ (a) Paris (b) The Moon (c) Jupiter (d) Mars” are not going to help anyone. Nor does including one obviously false choice, as students are not being challenged.
A good distractor can be something that students often get wrong. For example, many students are unclear about the difference between a model and a theory – and this isn’t helped by the incorrect use of the terms in some academic models and theories. Distractors can also be sensible ideas, but ones that don’t actually answer the question.
Shank, n.d. Origins and Purposes of Multiple Choice Tests. San Francisco State University: The Center for Teaching and Faculty Development. http://ctfd.sfsu.edu/teaching-practices/origins-and-purposes-of-multiple-choice-tests Accessed 15 May 2018.