Many of the comments here seem to be variations of, "I don't like editorials." I don't either, but their main point and recommendation seems pretty uncontrover

Many of the comments here seem to be variations of, "I don't like editorials." I don't either, but their main point and recommendation seems pretty uncontroversial: > Colleges need to improve online courses before they deploy them widely. Moreover, schools with high numbers of students needing remedial education should consider requiring at least some students to demonstrate success in traditional classes before allowing them to take online courses. They mention some studies (that I haven't read because I'm at work; I'm sure they're not bulletproof) that are broadly consistent with that recommendation. It would be great to have a truly randomized experiment, but that's unlikely to happen (although one could approximate a randomized experiment w/out too many ethical implications. For all I know the studies do that, but I don't know). But, as someone who almost never attended lectures as an undergrad, all of the same arguments about online lectures can be made (and I have made) about learning directly from the textbooks. Most people seem to find that difficult, so I wouldn't be totally shocked to learn that there is a tangible benefit to actually being in the same room as the instructor and asking for clarification at the moment of confusion. Disclaimer: I teach at a university. For the 250 student intro lecture I teach, I would be unsurprised if it could be taught just as well online, but I would want decent data on it before making up my mind. For the small classes I teach, I'd be surprised.