I have to wonder if flipped classrooms make teacher time in class more inefficient because the teacher's search costs of grouping like questions rises. Along the lines of Zen Scanner's earlier comments, if 5 students have the same question but are unable to stop the teacher in real-time to voice their question, how long does it take the teacher to first identify the students with similar questions and, second, actually address the underlying misunderstanding? It's not like students will line up based on the time in the video where their questions first arose.
I think flipped classrooms ignore the path dependency of learning. Learning, IMO, arises from an endogenous dialogue between the student and teacher, and flipped classrooms assume there is a fixed linear path for learning for all students. It sounds like the method could, if used incorrectly, actually become less adaptive to individual needs than traditional methods. I also echo DeeDee's concern that the model over-assumes teacher ability to effectively differentiate.
