Categorizing Learners: Bad Uses for Good Tools

by Kyle Skinner

Over the last couple of years, while becoming more invested in developing myself as a teacher, I’ve been asked to take the VARK inventory three or four times. VARK was presented to me as a learning style inventory, or a sort of personality quiz that tells you what kind of “learner” you are—visual, auditory, read/write, and/or kinesthetic. If you haven’t taken the VARK inventory before, you can take it for free here. Whether you’ve taken it or not, you’ve probably taken something like it before. Quizzes or inventories of learning styles like VARK, Kolb’s Learning Styles Inventory, or the Honey and Mumford Learning Styles Questionnaire are easy to find and seemingly ubiquitous. While they follow slightly different models from one another, they are all built to sort students into categories of learners. One study included a list of more than seventy available learning style models, and another even cites statements on an old version of our very own Yale Graduate Teaching Center website to demonstrate the extent to which the idea of learning styles has become popular in the world of pedagogy.

When asked to take these inventories in the context of teaching workshops, facilitators often justify their use of the quiz with vague allusions to the “matching hypothesis,” or the notion that learning outcomes can be improved by matching your presentation of course material to the style of learning that individual students prefer.

I have always been suspicious of VARK and learning inventories like it, but the last time I was asked to take it I got stuck on one particular question:

You are helping someone who wants to go to your airport, the center of town or railway station. You would:

-write down the directions.
-draw, or show her a map, or give her a map.
-go with her.
-tell her the directions.

One of the things I like about the VARK inventory is that you can select multiple answers to any question. But this question in particular (and a few other similar questions) makes me want to explain myself. Depending on whether the recipient of my directions is a beloved friend or a stranger, I might have different feelings about walking her to the airport. I might decide to write down directions instead of drawing a map if the directions would be simple and my piece of paper was small. Most importantly, I might give directions in different ways based on the navigational prowess of the recipient (my sister has such good directional memory that I could give her verbal directions once, but I would rather just put my mother in an Uber to save us both the inevitable frustration). The way I would answer this question would be dependent on context, not on my own personal learning preferences.

Even if the questionnaire did accurately diagnose my learning preference, I find myself skeptical that such diagnoses for my students would be particularly helpful to me as a teacher. When asked what their favorite days of class were over the course of a semester, I would be disappointed but not surprised were some of my more honest students to rank the lessons featuring active learning and hard work below the one day I screened a film. And I wouldn’t blame them—everyone needs an easy day every once in a while, but if curriculum were driven simply by what our students prefer to do, I think we’d see very little learning while YouTube would see a moderate increase in ad revenue.

My intuition seemed to pan out as I looked for evidence-based considerations of the usefulness of the matching hypothesis. A meta-analysis of relevant research revealed no reason to believe that the matching hypothesis is real. I far prefer to learn by reading silently to myself—but that’s because I’m impatient as a listener (as many close friends and former girlfriends would agree) and find that diagrams make me work too hard. But I could probably learn a lot more by becoming a better listener and analyzing diagrams. In fact, when I “match” my learning style, my intuition is that I get less out of it because I end up putting less effort into the learning.

I was feeling pretty misled about the VARK inventory and the like until I found this punnily named article, co-authored by a higher-ed consultant and Neil Fleming, the designer of the VARK questionnaire. I was surprised and delighted to find out that the questionnaire was never designed to be diagnostic, but rather was intended to be the starting point of useful conversations about meta-cognition that might help students themselves become better learners by thinking more about circumstances that aid or stifle learning:

“I sometimes believe that students and teachers invest more belief in VARK than it warrants. It is a beginning of a dialogue, not a measure of personality. It should be used strictly for learning, not for recreation or leisure. Some also confuse preferences with ability or strengths. You can like something, but be good at it or not good at it or any point between. VARK tells you about how you like to communicate. It tells you nothing about the quality of that communication.”

Using VARK as a diagnostic tool to determine what “kind of learners” our students are is a missed opportunity. Used to determine how we can match our teaching to our students’ learning styles, VARK encourages a one-sided teacher-centric classroom. For example, after finding out which learning modalities my students prefer, I can change my curriculum to match their preferences. Teachers should, of course, tailor their curricula to make sure they are meeting learners half-way, but providing information only in a student’s preferred learning modality means the student won’t get practice in learning in other styles. Instead of using VARK to dictate how we should teach or to inform our students that they should focus on particular methods of learning, the inventory could be used as the first step in a series of conversations with students about metacognition, ultimately helping them to develop their own notions of how to learn most effectively in the classroom or while studying on their own time.

There’s nothing wrong, I think, with the VARK inventory. Or any of the learning styles inventories you might find—but even the best tools are only useful when used well.


  1. Coffield, F., Moseley, D., Hall, E., & Ecclestone, K. (2004). Learning styles and pedagogy in post-16 learning. A systematic and critical review. London: Learning and Skills Research Centre.
  2. Fleming, Neil, and David Baume. “Learning Styles Again: VARKing up the Right Tree!” Educational Developments 7.4 (2006): 4-7. Web. 12 Jan. 2016. <>.
  3. Fleming, Neil. “Introduction to VARK.” VARK. VARK Learn Limited, n.d. Web. 12 Jan. 2016. <>.
  4. Fleming, Neil. “The VARK Questionnaire.” VARK. VARK Learn Limited, n.d. Web. 12 Jan. 2016. <>.
  5. Pashler, Harold, Mark Mcdaniel, Doug Rohrer, and Robert Bjork. “Learning Styles: Concepts and Evidence.” Psychological Science in the Public Interest 9.3 (2009): 105-19. Web. 19 Jan. 2016. <>.