But after thorough before-and-after testing of students taking the MITx physics class 8.MReVx (Mechanics Review) online, and similar testing of those taking the same class in its traditional form, Pritchard and his team found quite the contrary: The study showed that in the MITx course, “the amount learned is somewhat greater than in the traditional lecture-based course,” Pritchard says.

A second, more surprising finding, he says, is that those who were least prepared, as shown by their scores on pretests, “learn as well as everybody else.” That is, the amount of improvement seen “is no different for skillful people in the class” — including experienced physics teachers — “or students who were badly prepared. They all showed the same level of increase,” the study found.

The one type of class in which students learned even more effectively than in either online or traditional classes, the study found, was an approach called “interactive engagement pedagogy,” where students interact frequently in small groups to grapple with concepts and questions. Such “constructive engagement” in the classroom is something education reformers have long pushed for, Pritchard says, and is already used in many MIT classes.

Fiona Hollands, a senior researcher at Teachers College of Columbia University who was not involved in this study, says, “In my opinion, this study represents the most rigorous attempt to date to measure learning in a MOOC. This study provides an excellent demonstration of how learning in a MOOC, or in other types of courses, can be rigorously assessed. Applied to a broader population of students and a variety of educational settings, such investigations would provide valuable information about the relative effectiveness of different forms of educational delivery.”

I found "All cohorts learn equally," as intuitively plausible as "people of different heights grow at the same rate".

I think this is an important point of this study because one possible criticism of MOOCs is that they might only work for people who are already very high functioning and good at figuring stuff out themselves, and people who are more apt to struggle might be left completely in the dust (more so than in a live-contact situation). So the fact that this was not true is interesting.

Mazur’s reinvention of the course drops the lecture model and deeply engages students in the learning/teaching endeavor. It starts from his view of education as a two-step process: information transfer, and then making sense of and assimilating that information. “In the standard approach, the emphasis in class is on the first, and the second is left to the student on his or her own, outside of the classroom,” he says. “If you think about this rationally, you have to flip that, and put the first one outside the classroom, and the second inside. So I began to ask my students to read my lecture notes before class, and then tell me what questions they have [ordinarily, using the course’s website], and when we meet, we discuss those questions.”

Thus Mazur begins a class with a student-sourced question, then asks students to think the problem through and commit to an answer, which each records using a handheld device (smartphones work fine), and which a central computer statistically compiles, without displaying the overall tally. If between 30 and 70 percent of the class gets the correct answer (Mazur seeks controversy), he moves on to peer instruction. Students find a neighbor with a different answer and make a case for their own response. Each tries to convince the other. During the ensuing chaos, Mazur circulates through the room, eavesdropping on the conversations. He listens especially to incorrect reasoning, so “I can re-sensitize myself to the difficulties beginning learners face.” After two or three minutes, the students vote again, and typically the percentage of correct answers dramatically improves. Then the cycle repeats.

At M.I.T., two introductory courses are still required — classical mechanics and electromagnetism — but today they meet in high-tech classrooms, where about 80 students sit at 13 round tables equipped with networked computers.

Instead of blackboards, the walls are covered with white boards and huge display screens. Circulating with a team of teaching assistants, the professor makes brief presentations of general principles and engages the students as they work out related concepts in small groups.

Teachers and students conduct experiments together. The room buzzes. Conferring with tablemates, calling out questions and jumping up to write formulas on the white boards are all encouraged.

“There was a long tradition that what it meant to teach was to give a really well-prepared lecture,” said Peter Dourmashkin, a senior lecturer in physics at M.I.T. and a strong proponent of the new method. “It was the students’ job to figure it out.”

The problem, say Dr. Dourmashkin and others in the department, is that a lot of students had trouble doing that. The failure rate for those lecture courses, even those taught by the most mesmerizing teachers, was typically 10 percent to 12 percent. Now, it has dropped to 4 percent.