The flipped learning craze hit education a few years ago. Yet, new research raises questions over its impact on learning. The researchers have proposed an updated approach based on conditions that make flipped learning effective. It is called the 4F model.

“Flipped learning” belongs to a plethora of educational interventions that carry the same label, but vary substantially in how they are implemented in practice.

Not all such variations are equally effective for student learning.

Flipped learning typically comprises two phases:

  • A “passive” pre-class phase where students typically study pre-recorded learning content.
  • An “active” in-class phase where instructors engage students in problem-solving, discussion, and dialogue.

Proponents of flipped learning argue that moving passive content pre-lesson frees up space for more active learning in class, leading to better learning outcomes.

At least, theoretically, that is the claim. Practically however, and contrary to popular belief, our Learning Sciences and Higher Education research team at ETH Zurich, in collaboration with researchers from the Melbourne Graduate School of Education at University of Melbourne, did not find evidence for this claim (Kapur et al, 2022).

By examining data from 43,000 students (with a mean age of 18) across 173 studies, we found a 60% chance that a student picked randomly from a flipped learning implementation would have better learning outcomes than a randomly chosen student exposed to traditional instruction.

These learning outcomes were particularly better for:

  • School-age children compared to university students.
  • Students in developing countries compared to developed countries
  • Students exposed to shorter interventions (less than one month) compared to longer ones (e.g. a term).

All of this sounds promising, but because there was substantial variation in how both flipped and traditional learning was actually carried out in practice, we found it difficult to interpret these effects. For one, we did not know what was responsible for these effects. Therefore, we examined specifically how much of both active and passive learning were carried out during flipped learning.

To our surprise, flipped learning had the greatest impact on learning outcomes when the in-class phase included a short lecture. Conversely, we did not find much of an effect of active learning on flipped classrooms.

This means that the benefits of flipped learning were likely coming from passive and repeated exposure to learning content and less so from active learning.

On the other hand, when in-class instruction in traditional classrooms was designed well to include active learning, the relative efficacy of flipped learning diminished.

It seems, therefore, that the benefits of flipped learning could be replicated simply by just teaching well in traditional classrooms.

However, if one must flip, we combine our findings with recent advances in the learning sciences, specifically research on productive failure and problem-solving prior to instruction, to derive an alternative model for flipping.

 

The 4F model

We call our alternative model the 4F model: “Fail, Flip, Fix, and Feed.”

  • Fail provides opportunities for the instructor and the student to diagnose, check, and understand what was and was not understood via problem-solving.
  • Flip is consistent with the logic of flipped learning and is meant to provide pre-exposure to the ideas in the upcoming lesson (e.g. via a pre-class reading).
  • Fix provides in-class opportunities to explore student misconceptions and re-engage in learning the ideas via a traditional lecture.
  • Feed recommends providing formative feedback to students about their level of understanding, a missed opportunity in the typical implementation of flipped classes.

Altogether, our analysis forces us to rethink the over-enthusiasm for flipped learning and be cautious about the conditions that are needed to make it work (and that are often not there).

Yes, flipped learning can indeed be many things, but combined with advances in learning science our research points to a more effective, holistic model.

We invite educators and researchers to consider adopting and testing the 4F model – Fail, Flip, Fix, and Feed – in ways that advance science as well as practice.

  • Manu Kapur and Tanmay Sinha are from the Learning Sciences and Higher Education research team at ETH Zürich in Switzerland.
  • John Hattie and Irina Grossman are from the Melbourne Graduate School of Education at the University of Melbourne in Australia.

 

The research paper

Flipped learning is an instructional method that has gained substantial interest and traction among educators and policy-makers worldwide. However, current levels of enthusiasm for flipped learning are not commensurate with and far exceed the vast variability of scientific evidence in its favour.

The researchers conducted a meta-analysis of 173 studies. Based on their results, the researchers propose a more specific model for flipping – Fail, Flip, Fix, and Feed – whereby students are asked to first engage in generating solutions to novel problems even if they fail to generate the correct solutions, before receiving instructions.