tag:blogger.com,1999:blog-30114385.post3160869253874629597..comments2023-12-20T08:35:04.633+00:00Comments on Lingua Frankly: More Maths for MOOCs!Titchhttp://www.blogger.com/profile/03003350618976942468noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-30114385.post-72546260597013356072013-04-05T11:53:49.236+01:002013-04-05T11:53:49.236+01:00You would need more data than we have to calculate...You would need more data than we have to calculate it.<br /><br />And as for pedagogical value, there may arguably be more value in giving than receiving feedback, but receiving feedback is far from valueless.<br /><br />But even setting that aside and assuming that receiving feedback is of zero direct utility, there is still the issue of student engagement and motivation.<br /><br />There is nothing as demotivating as talking when there's nobody listening. (Something I think every classroom teacher can sympathise with!)Titchhttps://www.blogger.com/profile/03003350618976942468noreply@blogger.comtag:blogger.com,1999:blog-30114385.post-45611912146655963432013-04-05T10:48:42.072+01:002013-04-05T10:48:42.072+01:00I think it would be fairly easy to calculate the p...I think it would be fairly easy to calculate the percentage of people who get less than 1 comments if we assume a normal distribution. (I may have a go later)<br /><br />However I suspect the pedagogical value is more in the MAKING of comments than in receiving them.Johnhttp://www.feeldothink.orgnoreply@blogger.comtag:blogger.com,1999:blog-30114385.post-68121620932619029422013-04-05T09:54:15.223+01:002013-04-05T09:54:15.223+01:00That said, I suspect you're right about networ...That said, I suspect you're right about network effect resulting in a Zipf distribution in practice, but I was reacting to Martin Weller's words "at random" and taking them literally.<br /><br />But Zipf's law, if applicable, only makes matters worse, because it takes a system that already offers the possibility of unfair results and magnifies them.<br /><br />(However, if we were all to follow Martin's example, that wouldn't preclude going out and selecting additional blog posts based on personal preference, inter-blog linking etc, but the resultant distribution would be normal+Zipf, and I don't believe either would entire eliminate the effects of the other... I suspect we'd see something that can on a Zipf-with-a-hump....)<br /><br />And this all reinforces my determination to become an expert on statistics over the next year... it's a skill that's too little valued at the moment. (I don't believe anyone should leave university without knowing more stats than I currently know.)Titchhttps://www.blogger.com/profile/03003350618976942468noreply@blogger.comtag:blogger.com,1999:blog-30114385.post-75378398403141333472013-04-05T09:45:07.465+01:002013-04-05T09:45:07.465+01:00I was intending to turn my birthday example on its...I was intending to turn my birthday example on its head and include a reference to the classic <a href="http://en.wikipedia.org/wiki/Birthday_problem" rel="nofollow">birthday problem</a>, but I was interrupted by having to give a Skype English lesson and I lost my train of thought.<br /><br />But the main point is that regardless of what shape the distribution comes out at, the distribution includes a non-zero probability that someone gets no comments, which is unfair, unsustainable and just plain unacceptable in my book -- if this is "education", we need to do our utmost to give everyone an equitable share of the attention.Titchhttps://www.blogger.com/profile/03003350618976942468noreply@blogger.comtag:blogger.com,1999:blog-30114385.post-14498528898826713062013-04-05T08:45:04.650+01:002013-04-05T08:45:04.650+01:00on second thoughts I think I mean a normal distrib...on second thoughts I think I mean a normal distribution with the mean at three, which then gets distorted into a zipf shape by network effects.<br /><br />Please somebody else do the maths!Johnhttp://www.feeldothink.orgnoreply@blogger.comtag:blogger.com,1999:blog-30114385.post-43005764323989436112013-04-05T08:31:13.909+01:002013-04-05T08:31:13.909+01:00Niall, I'm enjoying your blog but not really s...Niall, I'm enjoying your blog but not really sure about your mathematics. If there are 365 blog writers and they distribute their (3*365=) 1095 comments at random I'd expect a zipf dsitribution: lots of posts with a small number of comments and a small number with lots. I think network effects will amplify this to exaggerate the importance of popular bloggers. The same as what happens with the internet as a whole.jhttps://www.blogger.com/profile/16596220450966683144noreply@blogger.com