University administrations are big fans of instruction over the Internet, because it is a big money maker. There is virtually no limit to the number of students in a class, one does not need a class room, and the class can be repeated at will once the initial investment is made. Faculty are less enthralled, because they think the classroom interaction is crucial, they fear of becoming expendable and they get sucked into a technology they are not familiar with. But, of course, one should first have an idea whether this is a teaching technology that works as well as the standard classroom.
David N. Figlio, Mark Rush and Lu Yin have performed experiments with a microeconomics class, where students were randomly assigned to Internet or classroom lectures. It turns out live learning works slightly better, in particular for Hispanics, male and otherwise low achieving students. At least for the latter two groups, from my experience these are the once that have a lower attendance record, and it looks like watching lectures when it best suits you is not an advantage, it seems rather like people postpone indefinitely and learn less. Unfortunately, the study says nothing about the on-line viewing habits. It would be interesting to see whether on-line instruction encourages absenteeism or tardiness in covering classes, something live lectures clearly prevent.
With the considerable influx of students in US public universities, whose resources keep decreasing, teaching on-line has been at the forefront of solutions to prevent overcrowding. While it does not seem to be a better solution in terms of learning, it certainly beats starting to offer classes at 6:30am.