Pashler study claims learning style models lack validity

The politics of research

Be wary of the way researchers present their conclusions as they are always political. Here, for example is an excellent YouTube presentation on the various types of research design and the paradigms they represent, as well as a related print source.

In the traditional corner, you have the quantitative experts, or what some would refer to as true scientific inquiry. They tend to assume that only experimental objectives and value free assumptions have practical utility as only those methods are able to answer “what” questions.

In the other corner, what I will refer to as the progressive corner,  you have the qualitative experts. They prefer questions that result in “understanding” and answers to questions that ask “why” and “how.” For example, qualitative or action researchers might want to understand how a student’s environment and cultural diversity impact on the way they learn — their learning styles — and thus their academic outcomes.  

Then, somewhere in the middle, there are researchers like myself who prefer to consider both progressive and traditional research approaches when doing a review of the literature, not either/or.

Pashler et al claim learning style models lack validity

And, so it was with great interest that I recently paid a visit to a parent forum I like to read from time to time, called Fixing Our Schools. Their top entry, dated July 21st, 2010, caught my eye. It was about a research study, published in a journal called the Psychological Science in the Public Interest sometime in 2009. It’s conclusion stated that:

“The contrast between the enormous popularity of the learning-styles approach within education and the lack of credible evidence for its utility is, in our opinion, striking and disturbing. If classification of students’ learning styles has practical utility, it remains to be demonstrated.” (My highlighting.)

Written by some distinguished scholars — Harold Pashler, Mark McDaniel, Doug Rohrer and Robert Bjork — I would certainly concur that the “learning styles” approach has become far too commercialized and clearly an educational bandwagon for much too long. However, as far as not being able to demonstrate the practical utility of using learning styles to guide instruction, I think we have to be very careful not to throw the baby out with the bath water. 

In other words, it’s not helpful for researchers to set up a straw man with rigid objective criteria — which they can then successfully knock down — which is exactly what Pashler et al did.

What research design and learning styles models did Pashler et al use?

First, they say they evaluated research studies that showed a clear statistical correlation between teaching that could link learning styles classifications to positive learning outcomes. Then, they looked for learning style models that related to that strict objective criteria in spite of the fact that the models were usually used in a subjective context.

For example, they used the Dunn and Dunn approach, which attempts to match learning styles with instructional activities, as well as the Kolb Inventory and the Myers Briggs Type Indicator. In the latter two cases, both are only concerned with internal states, not matching learning styles to instruction. In fact, the Myers Briggs is a personality type model that I used extensively when I was in private practice and meant only to reflect how someone prefers to make decisions, often with respect to career choice.

Lastly, Pashler et al use the conclusions of their study to recommend that traditional teacher-oriented “teach and then test” approaches should be used because they are more objectively scientific. The final sentence in their abstract states, for instance that “further research on the use of learning styles assessment in instruction may in some cases be warranted, but such research needs to be performed appropriately.”

But, therein is the influence of politics and research paradigms. What is appropriate to one researcher may not be so to another.

The crux of the matter

And so, we have a group of quantitative researchers drawing the conclusion that learning styles studies show no practical utility. Well, I disagree. It doesn’t have to be all or nothing. It doesn’t have to be quantitative method versus qualitative — left versus right or traditional versus progressive. And, its most definitely not up to Pashler et al to decide what method is most appropriate.

In other words, if learning style models help teachers and students in any way, they should be used, albeit not as an expensive bandwagon, rather as a guide for teachers on how to incorporate a variety of multi-sensory instructional techniques in their methods.

Now, back to my summer hiatus.