Thursday 2 November 2017

HE explained: The multiple chance test


The multiple choice test or "objective test" is much loved and much used by busy academics not simply because the questions and answers can be purchased, reused and marked automatically, but also because it appeals to an innate desire to measure achievement numerically.

Picture by Becris at freedigitalphotos.net
Detractors  (like my title this week) might think that it is akin to a lottery - 4 possible answers, one of which MUST be correct, and so a 1 in 4 chance of getting it right - 25% in the bag - high 5!

Only, look closely at the design of the multiple choice question, and you will see:

a) This IS the correct answer.
b) This is the opposite of the correct answer.
c) This looks almost correct, but for a different spelling or word or even double negative construction, designed to confuse.
d) The answer is always "The Archbishop of Canterbury", which is wrong and often a bizarre response to the actual question, apart from Ecclesiastical multiple choice exams where it might be correct or even a distractor.

So, the odds for the slightly prepared student get better.  Spot the obviously wrong answer and the odds shorten to 1 in 3.  Follow the logic and you could pass with 50% by sheer deduction, rather than subject knowledge.

And, don't get me started on feedback - I have known colleagues considering feedback to be the mark itself! Of course, neither questions nor answers can be shared, otherwise, they cannot be re-used.

If you really want to provide appropriate tests online or via computer then the multiple choice test is not the best answer.  There are a variety of other types of questions that can be deployed:

  • Multiple response - where all, or none of the answers may be correct.
  • Problem-based questions using diagrams, charts or pictures.
  • Short answer questions - where the exact words are not needed but enough to show knowledge of the subject.
  • Fill in the gaps....
Question difficulty can also be graded to allow more marks for "harder" questions.

My key advice, however, is to use a test instrument that actually measures the learning goal, not the one that minimises academic effort.



No comments:

Post a Comment