Hacker News Reader

Comment by pstuart | Hacker News Reader

San Jose State suspending its online offerings indicates that online only classes may not be all they're cracked up to be.

http://www.sfgate.com/news/article/San-Jose-State-suspends-o...

View on HN

Or that San Jose State did a crappy job of theirs. We'd need more than a short news article to determine which.

Theres no shortage of anecdotal reports by college professors complaining that higher than normal numbers of students fail/drop out of online courses. Heck, my own anecdotal experience of distance learning in college recently bears that out.

I'm told by someone who's taught one that the completion rates on free MOOCs are in the low single digit percents of people who signed up, if that. But, as he points out, this is really ok, as people are learning as much as they care to.

Yeah there is a difference when you actually pay for them, this is paid for. I've never completed a free one, but any online college course I've paid for you can bet I've completed spending several hours a week on and well over 16 hours or more leading up to mid-terms and finals.

if something is free, i bet many more people sign up to see what it's like. if it's not for them, they bail. a high percentage of dropout i would think is expected for something that's free. it's comparing apples and oranges...

Dropout numbers are a pointless metric on moocs. I've lost count on how many moocs I have signed up for, without any intention of completing them.

It's just an example of people trying bring in old metrics, on new things they don't understand.

Same here. I've learned a ton from the moocs I've signed up for and not completed.

The plural of anecdote is not data.

The plural of anecdote is data.

Edit: I suppose I should post something useful here rather than just a snark reply to someone's snark reply. Anecdotes represent distinct points of data and, therefore, a bunch of anecdotes would then be a collection of data.

I'm not sure that's correct. To become useful for science, you also need to put in a lot of work to correct for bias, etc. Having a plural number of anecdotes all with sample bias is not necessarily data and any attempt to create a correlation will run into problems.

The plural of anecdote is anecdotes.

The plural of datum is data.

Datum != anecdote.

Data != anecdotes.

QED.

And of course, college professors are completely unbiased in evaluating MOOCs...

I welcome high-quality MOOCs. I think competition will improve in-person offerings, especially at smaller colleges. I grow tired of getting low-quality students from my prerequisite courses. Sometimes I feel like removing the prerequisite requirements for some of my courses. The quality of some of them are so low.

San José State University just suspended the MOOCs -- their library school masters program (for example) is still mostly all on-line/ distance education

The article isn't very clear, but I think this SJSU/EdX collaboration was an attempt by SJSU administrators to save money by doing a "flipped classroom" with MIT educational materials. Maybe it's not politically correct to say so, but an average failing grade by SJSU students on MIT curricula does not surprise me at all. SJSU only has a 47% six year graduation rate.

How many SJSU students can afford to go to 4 years of college full time?

19% part-time students.

https://bigfuture.collegeboard.org/college-university-search...

One counterpoint to this is daphne coller's claim that pass rate is about 70 percent for people who paid $50 for a special track that includes verification of identity and means to prevent cheating.

There will be failed starts here and there. It does not invalidate the model.

Reply on news.ycombinator.com