Teachers, Learning Styles, and Using Data to Drive Instruction

Everyone likes data that back their prejudices. Academics call it “confirmation bias.” It runs rife among U.S. Presidents, state governors, legislators, school district policymakers, and Moms and Dads. I include myself in the crowd. People with beliefs on one or the other side of an issue lean heavily on examples and evidence that supports their view of, say, gun control, dieting, the worth of alternative medicine or the two-shooter theory in the Kennedy assassination. Resisting confirmation bias and being open-minded, a process that is closer to sandpaper rather than a soft pillow, requires awareness of one’s beliefs, values and positions on issues. It is hard work and requires attention in what one chooses to read, listen to, and think because it is far easier to screen out or avoid contrary information. Convenience often trumps thinking. All of this is also true for teachers. Consider the issue of  data-driven instruction and learning styles.

Gurus, vendors, and policymakers urge teachers to use test data such as research results and experts’ lists of “best practices” in their daily lessons (see here and here). The record of actual teacher use of data in classrooms however, is, at best, spotty. What happens, then, when research evidence overwhelmingly goes against deeply-held teacher beliefs in “learning styles?”

The most recent and detailed reviews of the literature on learning styles reveals little support for providing materials that play to the auditory, visual, tactile, and other ways that students learn (see here and here). Yet teacher beliefs about the importance of differentiating instruction to meet students’ varied interests, attitudes, abilities, and “styles” continue to be unflagging (see here).

Making sense of the contradiction between using data to drive classroom decisions and the poverty of studies that support “learning style”

So many teachers and studies have claimed–the operative phrase is “research shows”–that learning styles exist and differentiating instruction to match varied “styles” will lead to higher academic achievement (see here and here). Yet meta-analyses of available research has found little concrete evidence of such linkages. Even more so, the research on student learning styles is often deeply flawed in both design and methodology. Researchers have concluded:

Our review of the literature disclosed ample evidence that children and adults will, if asked, express preferences about how they prefer information to be presented to them.

There is also plentiful evidence arguing that people differ in the degree to which they have some fairly specific aptitudes for different kinds of thinking and for processing different types of information. However, we found virtually no evidence for [causal links between styles and achievement]…. Al-

though the literature on learning styles is enormous, very few studies have even used an experimental methodology capable of testing the validity of learning styles applied to education. Moreover, of those that did use an appropriate method, several found results that flatly contradict the popular meshing hypothesis.

A less jargony and concise analysis of learning styles and connections to student outcomes can be found in psychologist Daniel Willingham’s Q & A on the subject (see here).

So what’s going on here? Are teacher beliefs so powerful as to overcome strong findings that challenge those very beliefs? The answer is, unsurprisingly, yes. Not only do teacher beliefs in learning styles trump evidence but similar tensions between beliefs and data-driven decisions occur around direct instruction, multiple intelligences, and holding students back for a semester or year and other practices. But teachers, of course, are not the only professionals to succumb to confirmation bias. Doctors, lawyers, software engineers–name the profession–have similar issues. In short, the practice is pervasive among professionals and average folk.

If such cognitive bias is rife among the highly and barely educated, where does that leave data-driven instructional decisions as a “best practice?” Such biases mean, at least to me, that for any reforms aimed at teaching and learning the very first step is to deal openly and directly with the varied beliefs (and assumptions) that teachers have about their content knowledge, how best to teach that content to the young, and how do children learn more and better. These beliefs influence teacher choices of daily activities, instructional materials, arranging classroom furniture, what subject-matter and skills to teach, and grouping of students.

Without attention to teacher beliefs, confirmation bias will continue to go unnoticed as it so often does among physicians, lawyers, entrepreneurs, CEOs, and software developers. And data-driven instruction will remain lofty rhetoric rather than classroom realities.