Saturday, September 10, 2011

Effective Questioning


In 2002 Trevor Kerry estimated that teachers ask on average 43.6 questions per classroom hour. Given this frequency it is perhaps not surprising that they don’t wait very long for an answer: Kathleen Cotton (1988)  estimated that the average time allowed a pupil to begin answering was less than one second.





Some of the purposes of questioning (after Cotton 1988) include:


  • To motivate learners
  • To assess learners
  • To revise previous learning
  • To nurture insights
  • To develop learning skills

"These purposes are generally pursued in the context of classroom recitation, defined as a series of teacher questions, each eliciting a student response and sometimes a teacher reaction to that response." (Cotton 1988). Students have to:

  • Pay attention to the question
  • Understand what the question is asking
  • Think of an answer
  • Articulate the answer.


Cotton's meta-analysis of research led to some general conclusions:


  • Asking questions improves learning
  • Frequent questions improves learning of facts but does not improve learning of complex material (some suggest it can make such learning worse)
  • Oral questions are more effective than written questions
  • Questions which focus attention on key features result in better comprehension



For years, before watching a video, I gave students a list of questions to be answered during the video. I assumed that this would encourage them to pay attention to the video and result in enhanced learning both for the things specifically targeted by the video but also for other information (because they are more focussed). Cotton (1988) seems to imply that this is effective for students who are older, cleverer and more motivated but that younger children and poorer readers will focus exclusively on the questions and may miss out on other material. Nevertheless, it still seems worthwhile.


Another technique I have used in class is the quick-fire quiz. This started when I was teaching a modular Science GCSE which was assessed using short fact-based questions from a question bank. Very quickly I became able to predict the questions so I started every class with a quiz using the same questions. My favourite one was 'What does background radiation come from?' to which the answer was 'the sun [1], the stars [1] and rocks [1] especially granite [1]'. I tried this with a  very low achieving class. One lad averaged 2/60 on these module tests. On the Radioactivity test he scored 6/60, the difference being the 4 marks for the background radiation question. My triumph was spiled only by the realisation that he had given the same answer (the sun, the stars and rocks especially granite) to several questions. 


Nevertheless, rote learning seemed to have a part to play. I persisted. Another class included a very able but very lazy lad (later he became a crack addict) who averaged about 15/60 on these tests because he did absolutely no revision. On the Radioactivity tests he scored 45/60 because I had forced him to do the revision. There are times when you can lead a horse to water and make him drink!


But "Should we be asking questions which require literal recall of text content and only very basic reasoning? Or ought we to be posing questions which call for speculative, inferential and evaluative thinking?" asks Cotton (1988). The research suggests that:
Lower cognitive level questions are better than high cognitive level questions at primary; at secondary it is better to use a mix of levels.
Teachers ask lower level questions to those students they perceive as less able.
If you as lower level questions you should make them easy enough to answer.
Lower level questions are better if you want to teach knowledge (ie lower level questioning for lower level cognitive aims: not a surprise!)

Teaching students to draw inferences results in higher learning gains.
For secondary students using significantly more than the 20% average of higher level questions produces greater learning.
For secondary students using 50% or more higher level questions gives:

  • Better behaviour
  • Longer answers with a greater number of complete sentences
  • A greater number of both relevant answers and relevant questions asked by students
  • A greater number of student-student interactions

Wait time



On average teachers allow students less than one second to answer. They give less time to those perceived as less able.


The best wait time for lower cognitive level questions is 3s. For higher level questions the longer the better.


If you wait longer than 3s this will give:

  • Better achievement
  • Better memory
  • A greater number of responses especially at a high cognitive level. This increase is greater for student reluctant to participate.
  • Longer responses more often backed up by better evidence
  • A greater number of unsolicited responses, student-student interactions and questions posed by students.
  • Teachers will listen to students more carefully and engage in discussions more often.
  • Teachers will expect more of students.
  • Teachers will ask more varied questions with a higher proportion at a higher cognitive level.

Given how important questions are it is worth asking how they can be made more effective. These ideas come from John Mason.
·        Use assertions rather than questions to control behaviour. This keeps questions pure for pedagogy.
·        Develop a questioning classroom by praising pupils for attempting answers and for changing their minds when necessary.
·        Don’t play ‘guess what I’m thinking’ but ask genuine questions. “Be genuinely interested  not only in what learners are thinking, but in how they are thinking, in what connections they are making and not making.”
·        Don’t ‘funnel’ down to the ‘correct’ answer by asking simpler and simpler questions. Many pupils know this game and will wait before answering until the teacher has done all the thinking work!
·        Scaffolding (using direct questions) is good at the start of learning but you need to fade the support away using increasingly indirect prompts so that pupils learn to think for themselves.
·        Learn how to wait a little longer by formulating an answer in your own mind while you are waitin-g for the pupil to respond.
·        Learn what a pupil sees as important in a problem by asking them to read it aloud and listening to the words they stress.
·        Ask meta-questions such as ‘Is this always, sometimes or never true?’ or "What is the same and what is different about …?"
·        Get pupils to classify information or problems and ask them to explain their classification system.
·        Get learners to make up their own questions.