3 ms·

> Another example is Coursera and the like. You have Ivy League schools recording their lectures. Well... if I can learn CS 101 from MIT for free... True! Abst

by Kalium 10y ago

> Another example is Coursera and the like. You have Ivy League schools recording their lectures. Well... if I can learn CS 101 from MIT for free... True! Abstractly, 100% perfectly true for everyone. However... most people don't really learn well in a non-interactive environment where they most be completely self-directed. So in practical terms, it's far from ideal for most learners. It's not a coincidence that non-completion rates in MOOCs are quite high. "I can watch recorded CS 101 lectures from MIT" is really not the same as "I learned CS 101 while attending MIT and being in an instructional environment".

inanutshellus 10y ago
Completion rates (as a percent of sign-ups, anyway) are rather irrelevant. People sign up for free classes, poke around, move on. But you could easily see a world where--regardless of drop-out rate--more people completed an MIT-taught Coursera class than actually took the "real" MIT class.
- Kalium 10y ago
  I can easily see such a world is possible, yes. That said, completion rates do matter. If completion rates remain so sharply low, it will suggest that perhaps MOOCs are not quite the perfect replacement for instructional institutions they are often cast as. After all, success rate is the difference between telling a randomly selected American kid "Anyone can become President!" and them actually doing so.
  - rokhayakebe 10y ago
    You can compare completion rate for a free course and a paid course based solely on how many students enrolled and how many completed the class. You will have to set some point in the course at which you begin to compare, say for example you will start to compare all students who have completed the first half or third of the class.
Eridrus 10y ago
Dunno why you would expect them to be equivalent, this doesn't put pressure on MIT any more than being able to video call a doctor doesn't put pressure on the world's best surgeons. Comparing an MIT MOOC to night school is much more accurate. Completion rates there suck too.
- Kalium 10y ago
  I, personally, would not expect them to be equivalent. Self-directed learning with no feedback is much more difficult and error-prone than a guided process in an instructional environment. I use the comparison because a number of people seem to believe that the substitution is flawless and advocate for policy on those grounds.

by Kalium 10y ago

inanutshellus 10y ago

Completion rates (as a percent of sign-ups, anyway) are rather irrelevant. People sign up for free classes, poke around, move on. But you could easily see a world where--regardless of drop-out rate--more people completed an MIT-taught Coursera class than actually took the "real" MIT class.

Kalium 10y ago
I can easily see such a world is possible, yes. That said, completion rates do matter. If completion rates remain so sharply low, it will suggest that perhaps MOOCs are not quite the perfect replacement for instructional institutions they are often cast as. After all, success rate is the difference between telling a randomly selected American kid "Anyone can become President!" and them actually doing so.
- rokhayakebe 10y ago
  You can compare completion rate for a free course and a paid course based solely on how many students enrolled and how many completed the class. You will have to set some point in the course at which you begin to compare, say for example you will start to compare all students who have completed the first half or third of the class.

Eridrus 10y ago

Dunno why you would expect them to be equivalent, this doesn't put pressure on MIT any more than being able to video call a doctor doesn't put pressure on the world's best surgeons. Comparing an MIT MOOC to night school is much more accurate. Completion rates there suck too.

Kalium 10y ago
I, personally, would not expect them to be equivalent. Self-directed learning with no feedback is much more difficult and error-prone than a guided process in an instructional environment. I use the comparison because a number of people seem to believe that the substitution is flawless and advocate for policy on those grounds.