Thursday, 9 June 2011

Is it time to rethink the way university lectures are delivered?

A more interactive, discussion- and quiz-based style of university teaching brings dramatic benefits to science learning, according to a new study. The interactive approach takes its inspiration from psychologist Anders Ericsson's theory of "deliberate practice", a highly motivated and thorough form of learning.

Louis Deslauriers, Ellen Schelew and Carl Wieman parachuted into a physics course on week 12 and for half the year group (271 students) took over their three hours of lectures that week devoted to electromagnetic waves. A control group of 267 students were lectured by their usual, highly rated and energetic teacher following a conventional format (i.e. the students mostly sat and listened while he lectured). Both groups were set the same learning objectives.

Before the intervention, both groups had spent eleven weeks on the same course, albeit with different lecturers, and they were matched on mid-term exam performance and their engagement with, and attitudes to, class.

For the crucial week 12 lectures, the intervention students were led by Deslauriers and Schelew (both of whom have fairly limited teaching experience) and took part in a series of discussions in small groups, group tasks, quizzes on pre-class reading, clicker questions (each student answers questions using an electronic device that feeds their answers back to the teacher), and instructor feedback. There was no formal lecturing. The aim, according to the authors, was:
"...to have the students spend all their time in class engaged in deliberate practice at 'thinking scientifically' in the form of making and testing predictions and arguments about the relevant topics, solving problems, and critiquing their own reasoning and that of others."
The control group students had their usual lectures, covering the same material as the intervention students and they were given the same pre-class reading.

Student engagement (measured by trained observers) and attendance in the control group was unchanged in week 12 compared with earlier weeks. In the intervention group, attendance rose by 20 per cent and engagement nearly doubled.

In the first class after week 12, both groups were tested on what they'd learned in the previous week about electromagnetic waves. Also, two days before the test, students in both classes were emailed all the materials used by the intervention group: the clicker questions, group tasks and their solutions.

The results on the test were striking. The intervention group averaged 74 per cent correct, compared with 41 per cent correct in the control group. Factoring out the performance that could be achieved purely through guessing, the researchers said this meant the intervention group had performed twice as well as controls (the effect size was 2.5 standard deviations). Student feedback on the intervention was also overwhelmingly positive: 90 per cent of students said they'd enjoyed the interactive technique.

The researchers dismissed the idea that their findings could be explained by the Hawthorne Effect (i.e. a mere effect of novelty or of being observed). "While this experiment is introducing change in the student experience in one particular course (3 total hours per week) it provides little incremental novelty to their overall daily educational experience," they said.

The researchers' conclusion was upbeat: "We show that use of deliberate practice teaching strategies can improve both learning and engagement in a large laboratory physics course as compared with what was obtained with the lecture method ... This result is likely to generalise to a variety of postsecondary courses."

For a more critical analysis of this research, check out this NYT's article by Benedict Carey (thanks to an anonymous commenter for this link).
_________________________________

ResearchBlogging.orgDeslauriers, L., Schelew, E., and Wieman, C. (2011). Improved Learning in a Large-Enrollment Physics Class. Science, 332 (6031), 862-864 DOI: 10.1126/science.1201783

This post was written by Christian Jarrett for the BPS Research Digest.

13 comments:

  1. Time to get those mad fools at Festival of the Spoken Nerd to do some INSET days for teachers maybe?

    ReplyDelete
  2. Over the past 10 years - as a result of the evaluation I do after every training session - I've done less and less 'presenting' and more and more practice and discussion and trying to introduce novelty exercises throughout.
    The young (and old) generation are not passive listeners and crave interaction after a maximum 5 minutes.
    Its more fun for me too I admit!

    ReplyDelete
  3. Anonymous7:52 pm

    "I hear and I forget. I see and I believe. I do and I understand."

    - Confucius (551-479 BC)

    ReplyDelete
  4. I've been in teaching for 20 years (Sports, to Montessori to Math) and IT training for 15. My conclusion overall is that lecture is worthless. Practice is ALL of learning (+/- 5%).

    The primary value of lecture is to provide status for the instructor. Learning happens at the doing phase.

    My preferred lab/lecture split in training courses is now 90/10. But I can't manage that most of the time, as I have to use lecture to build credibility, and give the students what they're expecting.

    Also...my experience teaching pre-schoolers to Ph.D.s is that skill level makes a difference. Experts are more able to handle lecture than non-experts.

    ReplyDelete
  5. Anonymous10:50 pm

    The study is flawed, for reasons described in this NY times article:

    http://www.nytimes.com/2011/05/13/science/13teach.html

    ReplyDelete
  6. Anonymous1:17 am

    Because a study has flaws doesn't mean the results and conclusions aren't necessarily correct. It just means that tighter studies need to be performed, in other classes and other settings, and with the study authors not doing the teaching. Of course, even if the study had no discernible flaws, it would still need to be replicated. In either case, further studies are called for to find out how pervasive the effect is, and how best to extend these techniques more widely (assuming the effect is indeed replicated).

    As someone with multiple graduate degrees who has also done extensive formal teaching in a variety of settings, the results are overwhelmingly consistent with my own experiences. These researchers are on to something, and dismissing this study is a great way to help maintain the current dysfunction and grossly inefficient educational system we have.

    ReplyDelete
  7. Perhaps we should not cater to those with only a 5 minute attention span. College is not for everyone.

    ReplyDelete
  8. It seems that there are quite a few potential reasons why there are differences - some further experimental studies with more rigorous control should be carried out to back up some of the claims I think. But even if it makes little difference in terms of educational attainment, teachers should be trying to make their classes engaging, and it seems almost a 'no-brainer' that an engaging class is one in which you'll learn

    ReplyDelete
  9. Anonymous3:29 am

    Part of the problem is that few students have been taught how to learn from lectures. That kind of learning is a skill that has been neglected in lower schools due to a prejudice against lectures, which work quite well for many things (provided the students know how to use them).

    In my college teaching, I often encountered students who had "learned by doing" so often that they had figured out how to do, but not learn from it. Just filling time with pointless activity.

    When students learn how to learn, every method works.

    ReplyDelete
  10. Anonymous2:56 pm

    I am currently working and a student of Psychology. I do agree that interaction during lectures definitely play an important role and I feel it is more effective in the longer term compared to lectures-style teaching. 1. When we are to discuss ideas with peers, we will be forced to read up to, primarily to prevent ourselves from becoming embarrassed, and collaterally and more importantly in the academic sense, remember the information. 2. When we interact and exchange ideas, the discussions will impress students more and hence help us remember better. Hope this is helpful :)

    ReplyDelete
  11. Anonymous2:56 pm

    I am currently working and a student of Psychology. I do agree that interaction during lectures definitely play an important role and I feel it is more effective in the longer term compared to lectures-style teaching. 1. When we are to discuss ideas with peers, we will be forced to read up to, primarily to prevent ourselves from becoming embarrassed, and collaterally and more importantly in the academic sense, remember the information. 2. When we interact and exchange ideas, the discussions will impress students more and hence help us remember better. Hope this is helpful :)

    ReplyDelete
  12. I thought about doing this on a class i'll be giving on the next semester, and it just conviced me of doing it, i will write a little program where all my students will use clients that will show up questions and i will have a server reading those question.
    I'll be teaching Python programming, so i think this increases the chance of the technique being successful. I'm doing it because i remember a teacher i had that taught me World War 2 history and i really enjoyed her class, because she used this fast paced question driven learning environment i actually fell in love with the topic.

    ReplyDelete
  13. well, what about the costs? lecturers know the benefits of interesting examples, case studies, lab sessions but its just easier just to give a powerpoint lecture than prepare all the material, conduct the activity and evaluate it.

    ReplyDelete

Note: only a member of this blog may post a comment.

Google+