Preparing exam questions

I’ve been thinking about a question (based on a rather natural anxiety) that some of my students have asked me: “What will the questions in the final written exam look like?” and my spontaneous response was that I wouldn’t answer this question, or at least that I would not give a precise and concrete answer to this question. The reason is that I think that it’s important for students not to have a too narrow representation of what I want them to know and be able to do. Otherwise they will try to find out what they think I want them to know. If I gave them examples of questions that will be asked in the exam, then, I fear, they will prepare themselves too much and too narrowly to answer these questions… And I really want my exam questions to help me test their understanding and ability to flexibly use and apply the knowledge they have acquired/constructed throughout the semester… So the questions need to be NEW and unsuspected… It should not be possible to give a reasonable answer if you have not developed a deep understanding of the concepts and ideas.

However, I’ve recently come across a manifesto stating that “There’s No Excuse for Trick Questions” and I was wondering whether the type of questions I’ve been asking so far could be seen as trick questions…

Let me briefly quote Tina Blue (http://teacherblue.homestead.com/trickquestions.html):

What possible pedagogical value can there be in trick questions on exams or quizzes?

Several young people I know, both in high school and in college, have run into teachers who introduce trick questions on tests in a deliberate attempt to stump the student, even when the student knows the material he is supposedly being tested on.

[…]

Other than making the teachers who design the exams feel clever and showing the students how powerless they really are, there is nothing to be gained by such test questions. The purposes of an exam are (1) to give the student an opportunity to demonstrate his knowledge of the material and (2) to create a certain amount of pressure so that the student will review a body of material intensively enough to integrate it and understand it at a higher level.

It would be nice if it were possible to achieve these ends without the pressure of exams, but of course it isn’t, and so tests are–or at least they can be–quite valuable teaching tools.

But when a student has quite thoroughly mastered the material, there is no justification for trying to ask questions in a way that will make it close to impossible for him to demonstrate that mastery.

So I do agree that we should avoid trick questions, because their pedagogical value is zero. It sends out the wrong signal to a student who’s got mastery of a certain knowledge domain and still gets a rather negative feedback.

I do however also think that we need to ask meaningful and difficult questions… real questions, where there is no easy, simple and obvious answer.

So finally, I want to ask you, my readers, to tell me what you think… Should I give my students exemplar questions from the last sessions for instance, so that they better know what to expect and to be able to get this comforting feeling of being (better) prepared or should I encourage to develop their own inquiries, i.e., “What questions would you ask if you had to construct an exam?”? Please use the comments function below to share your thoughts…