I've started this thread because of a discussion that arose here: https://moodle.org/mod/forum/discuss.php?d=260673 and thought it deserved a thread of its own...
It's doubtful that we'll ever have a coherent "whole theoretical view" of learning. Some who are studying language acquisition are coming to the conclusion that it is a complex adaptive system, and therefore the traditional natural sciences approach to understanding is insufficient; one characteristic of complex adaptive systems is that they aren't possible to predict in any degree of certainty beyond the properties of the factors that constrain them.
For example, we know the properties of water, air, conduction, convection, etc., but at a certain point the system gets so complex that it becomes impossible to make accurate predictions about the weather. We can only talk about probabilities of weather events happening in relatively short timescales into the future (about 48 hours?). Now imagine that each of the water and air molecules has a mind of its own (agency) and can act against its own and others' interests "just because it wants to", and you begin to get an idea of how complex human learning and teaching can get. I don't think many cognitive scientists would argue with this view.
However, we can say that there are tendencies within constraints and that there are some strong predictors, i.e. activities, interactions, interventions, properties, attitudes, beliefs, etc., that tend to correlate with higher learning outcomes (Remembering that correlation isn't causation. Here's a humorous illustration: http://www.fastcodesign.com/3030529/infographic-of-the-day/hilarious-graphs-prove-that-correlation-isnt-causation ).
John Hattie has been getting a lot of attention for his work analysing correlations between learning interventions and conditions and learning outcomes. He's presented this list with effect sizes (1 = +1 average academic year's study, i.e. an additional year's gains in one academic year, and 0.4 is the baseline effect size, therefore anything above 0.4 is a good indicator) in one instance:
- Student expectations 1.44
- Response to intervention (i.e. teachers getting feedback about their teaching from learners) 1.07
- Teacher credibility 0.9
- Providing formative evaluation (i.e. descriptive feedback that also informs learners of what they should do next) 0.88
(i.e. self- and peer-observation of teaching by teachers followed by
critical reflection, AKA reflective practice) 0.88
- Classroom discussion 0.82
- Feedback (not sure how he distinguishes between this and number 5) 0.75
- Reciprocal teaching (AKA peer teaching, guided discovery, self-directed learning, etc.) 0.74
- Teacher-student relationships 0.72
- Spaced vs. massive practice (See: http://en.wikipedia.org/wiki/Spaced_repetition ) 0.71
- Metacognitive strategies (AKA "learning how to learn") 0.69
*For some reason, he's left out 2, 8, and 9 without any allusion to what they are.
Remember that these are indicators, and not necessarily causes, and because of the complex nature of learning and teaching, the effect sizes will vary greatly in individual cases. This list is intended as a very rough summary of what we should be looking to cultivate and augment in learners' learning experiences.
What do you notice about these top factors? What do they tend to have in common?