The idea sounds sensible: work out whether a student prefers to see, to hear or to do, and then teach them that way. Except that proving it takes more than the student doing well; it takes different types doing well under different methods. What was found?
A crossover that mostly is not there. The requirement is that each type succeed under its own method and lose ground under another, because pictures being good for everybody would explain a success just as well. Across 21 studies pooling 101 the average advantage was small, only about a quarter of the outcome measures showed the crossover pattern, and was low.
The idea sounds sensible: work out whether a student prefers to see, to hear or to do, and then teach them that way. The preference itself is real, and the claim about an advantage in learning is a separate claim.
The test of it is stricter than it seems. It is not enough for visual learners to learn well from pictures, because pictures may be good for everyone. demands that each type succeed under its own method and lose under another.
Across twenty-one studies, pooling a hundred and one effect sizes, the average advantage was small. But only about a quarter of the outcome measures showed the crossover pattern required, and the quality of the studies was low.
The conclusion is practical rather than principled: limited teaching resources are better spent on methods with strong evidence. The meta-analysis does not rule out every possible version of matching; it shows the broad application is not well supported.
wat. Only about a quarter of the learning measures in the meta-analysis showed the crossover needed to support matching a lesson to a "learning style".
wat wat
A personal preference is real, and it is not necessarily a recipe for learning better. Limited teaching resources are better spent on methods with strong evidence behind them.
How do we know?
The researchers pooled 101 effect sizes from 21 studies, and checked separately whether the results really showed a crossover between a student’s preference and the method of teaching.
What is still unknown
The studies were few and of low quality. The meta-analysis does not rule out every possible version of matching; it shows the broad application is not well supported.
Learn something?
Where to next
Do music lessons or chess make a child better at subjects nobody practiced?learning
Is sweet really tasted at the tip of your tongue?myths
Does living abroad really make a person more creative?learning