Last changed 17 April 2016 ............... Length about 1,000 words (7,000 bytes).
(Document started on 3 April 2016.) This is a WWW document maintained by Steve Draper, installed at http://www.psy.gla.ac.uk/~steve/myNewWave/discuss.html. You may copy it. How to refer to it.

Web site logical path: [www.psy.gla.ac.uk] [~steve] [myNewWave] [this page]

What do we know about student discussion?

By Steve Draper,   Department of Psychology,   University of Glasgow.

    What do we know about student discussion?

  1. Intuition, RDW, Newman: so this implies a research project, to see what evidence supports or opposes these views.
  2. We can induce it and demonstrate measured benefits reliably under certain carefully arranged conditions.
  3. But it has NOT been shown that it is worth sacrificing other learning time to do the discussion, only that adding discussion time brings benefits that would otherwise not happen.
  4. What measures show benefits? I.e. what are the benefits that have been shown?
  5. What do we know about the necessary preconditions for beneficial CI to occur? Mazur, Howe are identical for the preconds for beneficial CI: (The preconditions we know about:)
    1. Delayed effect
    2. 1. solo commitment first
    3. 2. Difference in initial opinions amongst the pair/group.
    4. 2b. Or doubts cf. Hunt.
    5. 3. Providing the topic from outside.
    6. 4. Fix the social bonding e.g. T imposes it on the group OR friends live together i.e. spontaneous = unplanned discussions.

    What do we NOT know about student discussion? What are the research questions to answer?

  6. What value is it? (is CI; or peer discussion in general). I.e. what kinds of value can be gained?
  7. How much peer discussion (CI) goes on, and why?
  8. How frequent is CI?
  9. When is it worth sacrificing other learning time to do peer discussion instead?
  10. What are all the types of peer discussion that occur? Cf. GPate's 3.
  11. ----
  12. How to create CI out of a given topic. Brain teasers? self-invented curiosity questions; both non-peer stimulated, and reflection which the enthusiast does anyway.
  13. How to turn CI into a habit, a learning skill.
  14. A borderline / spectrum.
    There is:
    1. Pure CI
    2. Do learning in pairs, where you ask each other
      1. Facts
      2. Reasons (for a fact or law)
      3. Real life examples
    3. Solo cramming not peers; but doing things to deepen understanding.
  15. x 
Under careful conditions, it has a large, valued, and reproducible effect.

Fits with many people's intuition.  RDW

Measures: a) deep learning; b) improved CT skill.
	But not often a clear boost to exam scores.

CI and catalytic effects.

What are all the types of peer discussion, and their values?
	CI.  Facts, often admin fact checking.  Checking the set task / problem.
	Teaching someone else (a close cousin to a brain teaser).
	Pooled free association; e.g. to link concept to personal experiences.

To get CI, you need usually a carefully designed question / conundrum (can't just say "discuss something" or "discuss conceptA".

Partly peer discussion is a social prompt for re-experssion (L-act 2), but so are writing assignments.
There are 3 such prompts: 
    
	
  1.  Full CI or debates
	
  2.  Teaching someone else
	
  3.  Writing (solo, though for some imagined audience)



Mazur, Howe are identical for the preconds for beneficial CI:
(The preconditions we know about:)
Delayed effect
1. solo commitment first
2. Difference in initial opinions amongst the pair/group.
2b. Or doubts cf. Hunt.
3. Providing the topic from outside.
4. Fix the social bonding e.g. T imposes it on the group OR friends live
together i.e. spontaneous = unplanned discussions.

fragments

There is a literature that shows that peer discussion (amongst students) under specific conditions significantly improves their conceptual understanding. Look at this page which gives an ancient argument for how discussion is essential for learning; and skim over Draper (2009), which gives some references on the literature on learner discussion. The Miyake and the Howe references are particularly relevant for seeing how discussion has been shown to stimulate learning.

What is less clear is when this is important, and whether it is best to leave it as unplanned (opportunitistic), or to organise it.

What might we like to believe about student discussion? (even if we can't say we "know" it)

Newman: put quotes from him here. 
0) Who Newman was; ref.
a) state his beliefs about not exams and professors
b) Reproduce his long Jane Austen-like sketch.

LTimmons

Actions to foment CI

LTimmons suggests that currently opportunity is the main driver. And in fact it may be quite rare (How many CI discussions per day? Few, I would guess).

See old TM "MOOCs" for notes on new software for having almost on-demand CI within (big) Moocs.

All RPC is a form of this, though not iterative: but it has the core Piagetian driver of you see a peer assert or do something quite differently from you, you notice the clash, it stimulates you.

Tweetchats. They are speed-CI. But not only do they work for SLH and her gang, but they should have the advantages that speed-X do in my expExp designs: make students feel it is not serious, so no penalty, and get them going.

Web site logical path: [www.psy.gla.ac.uk] [~steve] [myNewWave] [this page]
[Top of this page]