It is worth pointing out that this is not an isolated piece of research. It is such a strong effect that the author of that paper, Richard Clark, coined the term ‘mathemathantic’ for when particular teaching methods ‘kill’ learning and noted that instructional procedures where students are give control over learning tend to be mathemathantic. Paul Kirschner and Mirjam Neelen have an excellent blog post on the topic.
If you are still not convinced, then perhaps it might be worth sharing two papers that I just happened to have stumbled across recently.
The first paper is a study of students engaged in reading digital texts and print texts. Students preferred the digital texts and predicted that they had understood them better. However, testing demonstrated that students actually possessed a superior understanding of the print versions.
The second paper is a new one in Learning and Instruction and John Dunlosky, one of the authors, may be familiar to you from this piece in American Educator. Students were give choices of how to learn to solve certain probability problems. Few students chose the optimal strategy which was to study worked examples prior to attempting the problems.
I think there are two conclusions we may draw from this. The first is the fairly robust conclusion that giving students choices over how to learn will lead to less learning than a teacher choosing strategies according to research-informed principles. We may, of course, decide to give students choices for other reasons, such as motivation or variety, but we need to do this in the knowledge that we are trading this for effectiveness. The mantra at the heart of progressive education, that we should always follow the child’s interests, is deeply flawed. Teaching models such as Universal Design for Learning that are premised on student choice are hopelessly inadequate.
The second conclusion is more speculative. If students make the wrong choices then what about teachers and academics? If students are drawn towards less effective instructional strategies then might this explain why some of us hold on to beliefs about teaching that do not reflect the evidence? Perhaps constructivist teaching methods just feel right to people?