Cheating Or Mastering? | Hacker News Reader

Cheating Or Mastering?

rjlipton.wordpress.com

16 pointsjoeyespo11 comments

It's very simple, I think:

If you need these courses for credentialing, you have to lock them down more. Get a credit card, require a payment, and issue a token. Validate the name and address (and maybe store a hash of the SSN), and don't allow duplicates. Treat "cheating" as fraud, and act accordingly.

If you need these courses for lifelong learning, the problem doesn't exist. Let someone take it a million times if that's what they need to do to learn.

No, credentialling should be separated from learning. If I need to prove I can safely build a bridge, I should just be tested on that. If I want to learn how to build a bridge, my education should allow me to do whatever I need to do to master the subject.

A huge part of our problems in higher education and hiring have come because we've conflated the two.

tl;dr: I think that agrees with what I was saying, for some definitions of the word "separated".

The two activities credentialling and learning are strongly related, because testing can be a good way to give feedback during the learning process, in addition to being the only scalable way to filter masters from novices. As far as I can tell, apprenticeships and co-ops don't scale today. Interesting if they could.

I don't think it harms Udacity to run both - a free tier for people to get their feet wet in CS and explore through a lifetime of learning, and a pro tier that gives the equivalent of a Master of Science with certificate (ie, classes only, no thesis). The free tier allows unlimited retries, has no deadlines, and encourages self-learning. But the pro tier separates the wheat from the chaff a bit, is strict on identity, is strict on plagiarism, gives one-try test grades, and has a minimum bar for certification.

You could probably make the divide between free and pro in a much different manner than I just did. I think I gave a pretty old-school university way of thinking about the problem. Of course they are not confined to think this way.

The materials for both tiers could be very close to each other, to the point where it would be wasteful to separate instruction. That's why liberal education and professional education get smushed together into higher learning today. Maybe Udacity can separate out these two in an interesting way.

I have little doubt that the problems discussed actually exist, but the biggest reason for the difference is simple economics. Anything free, or at extremely low cost, is over-consumed; hence the many sign-us for the course. But completing a course is NOT free, or usually even cheap, it takes a great deal of time and effort for any but the simplest courses; therefore most end up dropping out when it stops being free.

As for the "cheating versus mastery" part, it depends on why and exactly how he is working it. Simply taking the course over and over could be either; signing up for multiple, simultaneous sessions and picking best scoring sessions is pretty obviously cheating.

But cheating whom, exactly? It is not like the mark matters at the end of the day. What matters is if the material has been learned, and if it takes several iterations to finally learn it fully, so be it.

>signing up for multiple, simultaneous sessions and picking best scoring sessions is pretty obviously cheating.

Why? Taking the course multiple times sequentially is potentially okay, but taking it multiple times in parallel is "obviously cheating"? What's the difference, assuming it is always the same student taking the test?

Isn't practice something to be encouraged, and not punished?

Practice is indeed something to be encouraged. Practicing to the test is no better than teaching to the test though. It's fine if there are no outside consequences like job offers, certificates with a score behind them etc. It is cheating though.

Its cheating because its against the rules.

The question is as the designer of the rules should you design that to be against the rules?

Practicing to the test is a seperate question and a function of how useful you make the test

Another interesting question is jobs, somewhere once (I don't have time to find a good link) there was talk that the business model behind Udacity was that it could sell more accurate info on students to employers than colleges can

for instance they can track who finishes stuff early vs who waits until the last minute to turn something in

At this stage, I don't think it matters if Bob cheated because Udacity can't even verify Bob is actually Bob. It does seem strange that the post doesn't talk about the logical next step, physical testing centers that can verify identity, especially since Udacity already announced their first partnership (http://udacity.blogspot.com/2012/06/udacity-in-partnership-w...).

As we move into a new world of education, our goal should absolutely be mastery. Trying to group students by arbitrary aspects like "age" don't help education. Forcing students into more advanced topics before they are ready or simply telling them "You suck. give up", isnt't helping, either.

I really like the schools the Khan Academy is working with that use the videos to describe and drill on the topic at home and classroom time to work through the problem sets where the teacher and other students can help.

Combining the mastery of this article with the Khan model means any two students may not be working on the same problems or watching the same videos, but they are all working at the level they understand and are able to progress as a result.

been a while since I was in college, but nothing was better for studying than just taking practice tests,

if nothing else, MIT's OCW, and all these online classes create a vast library of practice tests, if I was in college now I would try and sign up for as many parallel classes as I could, just so I had practice tests to work through prior to my test for college

Reply on news.ycombinator.com