Slashdot Mirror

The evidence is phenomenally consistent that the online self-paced stuff works great for professional people who've mastered college-level skills in reading, writing, and math... but falls on its face for people who don't have that. For example, every attempt at getting the horde of people who need algebra remediation through online course has been a disaster. UDacity tried it at San Jose state and was suspended after one semester. Community colleges in Philadelphia tried it and concluded "The failure rates were so high that it seemed almost unethical to offer the option". So I highly doubt you can replace elementary/secondary schools with this method; at that level, most student need a personal face and hand-holding through the material, especially with technical stuff like using, interpreting, and debugging online resources in the first place.

http://www.angrymath.com/2013/06/online-remedial-courses-considered.html
http://www.angrymath.com/2013/...

I think there is a bit of selection bias here. Each course has pre-requisite knowledge, and I suspect a large part of the reason people can struggle with a course (and therefore need a remedial course) is because they don't have a solid grasp of the assumed knowledge. This is especially true for subjects like math, where all the different subdisciplines are inter-connected (e.g. consider how often log and trig laws turn up in calculus).

This problem arises partly because students are not held back a year if they perform poorly in one or two subjects, but are then assumed to have the same knowledge as everyone else in the year. The primary advantage of an online system is that the content can be tailored to each student. e.g. you can force a student to re-do a unit of math until they are able to pass without impacting their ability to take higher level units in other subjects.

Of course, this doesn't in any way negate the need for one-on-one time with remedial students. But by automating the more repetitive parts of the knowledge transfer process, it frees up the educator to spend more one-on-one time with struggling students.

There are also issues with the increased need for students to have self-motivation and drive, but I suspect it would be beneficial to imbue them with those traits from a young age anyway.

"Where they interview 0.00125% of the population"

This is the dumbest goddamn thing you can say about statistics.

You had me, up until you referenced how many articles Wikipedia has. That shows you don't understand how sample sizes work, and thus statistics pretty much at all. http://www.angrymath.com/2010/...

Passing high school math and science courses would be more than enough in this case.

Which we may assume the professor of physics at McGill university has done. So what's the qualification you have that he doesn't?

And anyone with the slightest understanding of statistics knows that you cannot take a 500 year non-random sample out of a 500m year data set and expect this miniscule slice to accurately reflect the whole.

The slightest understanding of stats you say....

http://www.angrymath.com/2010/...

Oops.

My impression is that the Facebook system, for all its corporate might, is a nigh baling-twine-and-duct-taped mishmash of servers with unpredictable sets of code in them. Partly this is due to my discovery that sometimes your post history fails to be included in a downloaded archive (link). At the time, it was working for some people but not for others. It had worked for me in the past, then stopped, then restarted later on. A Facebook worker in the comments actually asked around and said the feature was supposed to still be there, and someone should be fixing it. But how on earth would part of a downloaded archive just entirely go missing, including the link in the index page? The only way I can parse that is that different servers are running different code and features in unpredictable ways, sometimes changing for part of the client population and not others.

That is the dumbest goddamn thing you can say about statistics.

"This poll is of only 1000 people, though; your mileage may vary."

Hey look -- it's the dumbest goddamn thing you can say about statistics.

"Sample of 162 in 9.5 Million"

This is the dumbest goddamn thing you can say about statistics.

"Okay so you're talking about roughly six and a half billion people. As of the writing of this post, votevotevote.net's page says: 1050 VOTES have been received"

This Is The Dumbest Goddamn Thing You Can Say About Statistics.

"A current example would be why is there a picture of a floppy disk to save data? Would any 8th grader know what a floppy disk is? If not, how does that icon make any sense at all?"

No. I've found that even for current college students, that icon has no meaning.