Research School Network: Using Diagnostic Questions How can we get better at diagnosing gaps in understanding?

A common dilemma with assessing understanding using diagnostic questions is knowing how much time to commit to it during a lesson when you are conscious of keeping up with the scheme of work. Another aspect to address is the time commitment of creating the questions and how much of your planning is taken up doing so. It is important we focus on mastery of learning and a key part of this is not moving on too soon which relies on regular formative assessment. One of the easiest ways to do this is to ask students to answer a multichoice question on their whiteboards or hold up a number of fingers to show which answer they have chosen.

It is important that the teacher does not assume that the students have a thorough understanding if they select the right answer as there is a chance they would have guessed correctly but not necessarily chosen the answer based on their own knowledge. Following up a diagnostic question with some probing questions is essential to ensuring you uncover as many misconceptions as possible. It is just as important to question the students who were incorrect as it is to question the ones who are correct. Students should be able to articulate why they chose their answer, whether correct or incorrect.

Sometimes it can be beneficial to just focus on one high-quality diagnostic question before moving on to the next concept. Dylan Wiliam writes that ​“for the purpose of rapid assessment of student learning…a single well-chosen question can provide enough information to direct instruction in real-time, provided the item is chosen carefully”. It is important to realise that you don’t always have to give up a huge portion of the lesson to diagnostic questions for them to be effective.

When looking for a diagnostic question, you could start by looking at a bank of questions such as those on the website diagnosticquestions.com which has an ever growing database of questions for various subjects. If you write your own multi-choice question, you should ensure that any incorrect responses unveil a particular misconception which will help to inform your next teaching decisions.

Here is an example of one I have recently used in maths (from the website Eedi.com) for indices laws:

Each answer has been specially selected to draw out a particular misconception. In answer B it would appear that the student has confused squaring with multiplying. In answer C, the student has either forgotten to square three or has not realised that the square function applies to the whole bracket. In answer D, the student has squared the three but is perhaps unsure of how the squaring affects the algebraic variable. All three wrong answers encourage a rich discussion into the depth of understanding required to master this particular law of indices and allow the teacher to pre-empt any possible misconceptions.

There is an argument that by presenting incorrect answers as part of the choice could encourage and embed certain misconceptions. This is why it is so important to consider and discuss the incorrect answers and confront the misconception head on. Moreover, if a student confidently choses an incorrect answer and then subsequently receives immediate feedback into why it is wrong, they are more likely to experience the ​“hyper-correction effect” and have a greater likelihood of remembering the correct answer in the future. Andrew Bulter and Henry L Roediger III write that ​“feedback enhances the positive effects of taking a test and helps students correct their errors, thereby reducing the acquisition of misinformation”. The evidence is clear in that if you are going to use diagnostic questions, be sure to give feedback to the students so that all misconceptions are dealt with.

Zofia Reeves

Research School Associate