Teaching via the flipped classroom method fails to boost student performance, a major study suggests.

The popular approach ¨C in which students are introduced to learning material before a taught session and then spend class time engaging in problem-solving and discussion ¨C may also exacerbate achievement gaps between different groups of learners, according to a randomised control trial conducted at a US institution.

The results, published in a Massachusetts Institute of Technology discussion , pose a challenge to the growth of the flipped classroom model, which is used in more than half of US universities and widely around the world, and is growing in popularity.

For the experiment, involving 29 academics across 80 economics and mathematics classes at the United States Military Academy, half of the 1,328 students were taught via a traditional lecture method, and half were instructed using a flipped classroom model, watching video lectures before teaching sessions and working on problem-solving tasks in class.

³ÉÈËVRÊÓƵ

Student achievement was compared using an in-class quiz taken after three taught sessions, and then the course¡¯s final exam. In economics, there was no significant difference in performance between the two groups. In maths, students in the flipped classroom group did better in the in-class quiz, but their advantage had faded by the time of the final exam.

Examining the results in more detail, researchers found that the short-term improvement in maths performance was driven by better scores for white, male students ¨C the achievement gap between white and black or Hispanic students was 69 per cent larger than in the lecture group ¨C and those who were already high performers.

³ÉÈËVRÊÓƵ

And, while any overall improvement had disappeared by the time of the final exam, these achievement gaps persisted: the flipped classroom cohort had a 51 per cent larger achievement gap by race, and a 25 per cent bigger gap based on prior attainment.

Elizabeth Setren, Gunnar Myrdal assistant professor of economics at Tufts University and one of the authors of the study, said ¡°the key takeaway is that if educators are interested in implementing it in the classroom, they should pay careful attention to how it affects students across the skills distribution¡±.

¡°We've seen a huge burgeoning interest in lots of different ways to incorporate technology in higher education, and the flipped classroom is one of the more popular models that people are exploring...it¡¯s important to bring some evidence to bear on that,¡± Dr Setren told ³ÉÈËVRÊÓƵ.

Some proponents of the flipped classroom model argue that it helps students from the lower end of the performance distribution. However, that was not the case in this experiment. ¡°If educators are hoping it will be a solution, they need to pay careful attention and think about how they can modify it to support those students,¡± Dr Setren said.

³ÉÈËVRÊÓƵ

Dr Setren added that, when looking at what might explain the differences between maths and economics quiz results, ¡°we found that the maths faculty were more motivated and interested in participating in a flipped classroom model, whereas in?economics they preferred the standard classroom¡±.

Rather than taking a top-down approach and imposing a change to the pedagogical model, university leaders should think about allowing instructors to choose for themselves. ¡°Perhaps the most motivated will see the largest results,¡± Dr Setren said.

anna.mckie@timeshighereducation.com