Students have different “learning styles” is a myth

Is there any evidence to support the learning styles concept?
Yes there is a little, but experts on the topic like Harold Pashler and Doug Rohrer point out that most of this evidence is weak. Convincing evidence for learning styles would show that people of one preferred learning style learned better when taught material in their favored way, whereas a different group with a different preference learned the same material better when taught in their favored fashion. Yet surprisingly few studies of this format have produced supporting evidence for learning styles; far more evidence (such as this study) runs counter to the myth. What often happens is that both groups perform better when taught by one particular style. This makes sense because although each of us is unique, usually the most effective way for us to learn is based not on our individual preferences but on the nature of the material we’re being taught – just try learning French grammar pictorially, or learning geometry purely verbally.

Are there any other problems with the myth?
Oh yes! Another major problem is that there are so many different possible ways to describe people’s preferred learning styles. Indeed, a review published in 2004identified over 71 different styles mooted in the literature. As Paul Kirschner and Jeroen Merrienboer explained in their recent article on “urban legends” in education, if we view each learning style as dichotomous (e.g. visual vs. verbal) that means there are 2 to the power of 71 combinations of identified learning styles – more than the number of people alive on earth! What’s more, even if we accept a particular scheme for measuring learning styles, evidence shows that learning style questionnaires are unreliable and people’s self-reported preferences are poorly correlated with their actual performance. In other words, a person might think they learn better, say, visually rather than verbally, but their performance says otherwise! The fact is, the more accurate predictor for how well a person will fare in a math learning task, is most likely not the degree of match between their preferred learning style and the teaching style, but their past performance on math tests.