How to double your student exam scores (yes, really!)

On 13 May, Deslauriers et al (2011) published a paper in Science (impact factor 29.7) where they demonstrated the hugely beneficial effect of a teaching method called ‘deliberate practice’ (see chart below).

They employed this method for 3 × 50 mins in week 12 of a module – each of the previous 11 weeks had contained 3 × 50 mins of traditional lectures. When contrasted against a cohort that had received traditional lectures in week 12, the subsequent scores in a 15-20 minute MCQ class test were amazing… particularly as deliberate practice was taught by relatively inexperienced teaching assistants, whereas the traditional lectures were given by a highly experienced lecturer who often attained high student feedback.

What is deliberate practice?

It is an active learning approach, which uses the whole of the lecture time for problem-centred learning, and where the students frequently receive feedback.

How did Deslauriers et al implement deliberate practice in their teaching sessions?

One instructor and one assistant ran the class of 271 students. Before each session, the students were instructed to read a 3-4 page reading and answer a related on-line true-false quiz. At the start of the first teaching session, the students were told the rationale for the use of deliberate practice (and supporting research was cited).

A typical session schedule was:

An MCQ question was posed, and students discussed the answer with other students (2-3 mins).

The students gave their answers using ‘clicker’ handsets (like Optivote), and the instructor gave feedback to the whole cohort (4-6 mins).

This process was repeated with another 5 different MCQs.

Sometimes, an MCQ was presented twice in succession, presumably so that the effect of the instructor feedback and 2nd student discussion could be measured. Note that this seems to suggest that the instructor feedback did not always give the correct answer.

Twice in each teaching session, there was a small-group active learning task (6 mins), followed by instructor feedback (4-6 mins).

How long did it take to prepare each 50-minute teaching session?

Including the pre-session reading quiz, 20 hours at first, dropping to 10 hours for the 3rd session. But the authors estimate that this would fall to 5 hours with practice. Note that these times include piloting the materials with one or two students, and subsequent editing.

How was deliberate practice received by the students?

Deslauriers et al surveyed the students and found that they enjoyed it more and thought that they learnt more than in traditional lectures. Interestingly, students thought that they learnt much more in class than from the pre-reading with quiz, even though the majority (all?) of the new information came from the pre-reading.

What conclusions can be drawn?

An active learning approach with frequent feedback…

  • is better than a traditional lecture.
  • can be applied with large cohorts.
  • does not need many teaching assistants.
  • does not take a lot longer to prepare (when practiced).

Where can I find out more?

A hard copy of the Deslauriers et al paper is in Aton’s library (but is not available electronically yet). The supplementary material is available at


Deslauriers, L., E., Schelew, E. & Wieman, C. (2011) Improved Learning in a Large-Enrollment Physics Class. Science 332(6031): 862-864.

One Response to “How to double your student exam scores (yes, really!)”

  1. reddypa  on June 22nd, 2011

    This is really encouraging – and great to see it in such a key journal. I am planning two short final year u/g modules (5 weeks each) and I am going to try for a clicker distribution to all students and design the programme along these lines.