Interesting Links for 23-02-2012

[andrewducker]

• Science Finds a Better Way to Teach Science. (Lectures are pointless. I wish I was even slightly surprised)
  (tags: academia teaching research science )
• The British idea that ordinary people don't need to understand mathematics causes appalling problems
  (tags: mathematics economics uk )
• How the European Internet Rose Up Against ACTA. (Gives me a warm feeling inside.)
  (tags: copyright acta europe internet censorship )
• The myth of the eight-hour sleep - humans may be adapted for two 4-hour naps, with a 2-hour gap in the middle
  (tags: sleep )
• What NOT to do when relaunching the Conservatives in Northern Ireland
  (tags: uk politics conservatives northernireland )
• A week of picking on trans people
  (tags: media uk lgbt transgender )
• The Only 4 Monitors You Should Buy
  (tags: lcd technology monitors )
• The iPhone umbilical cord charger. If I had an iPhone, I'd have one.
  (tags: awesome iphone )
• How Bots Seized Control of My Pricing Strategy
  (tags: pricing ai amazon )
• Orange, T-Mobile ready to launch 4G together in UK this year
  (tags: orange t-mobile 4g mobile_phones uk )

Comments

[andrewducker]

Hmm... you over state the case here but it is not news that lectures are not the most effective way to get the point across. Unfortunately most of the more effective methods are very staff intensive. Given one person and a set amount of material to be got through, a standard lecture may be pretty much your only realistic choice.

"A University of Maryland study of undergraduates found that after a physics lecture by a well-regarded professor, almost no students could provide a specific answer to the question, “What was the lecture you just heard about?” A Kansas State University study found that after watching a video of a highly rated physics lecture, most students still incorrectly answered questions on the material. Wieman himself found that when he quizzed students about a fact he had presented 15 minutes earlier in a lecture, only 10 percent showed any sign of remembering it."

I'm sorry, but that's not a realistic way of doing _anything_. Given a 10% success rate, the answer is not to carry on doing what you're doing, because it's all you can do with one person. If the choices there are "Waste everyone's time for an hour" and "Do nothing" then "nothing" is less wasteful.

[steer.livejournal.com]

Yes but my essential point is that really is a straw man argument because that isn't the system as it is in any university I have ever taught in. You are assuming that this is an argument against lectures. It is not. It is an argument against lectures which are not followed up with worksheets, seminars, self study and exams. So, if you take the system you describe it is a bad one. That's why no university course I ever taught on uses it. Which is also why university courses which rely heavily on lectures do teach students things (and we know this because we set and mark exams to check it).

I wouldn't expect students to retain much from just the lecture. I would expect the students to retain the material from the lecture followed up by worksheets, self-study and practice exam questions. In many universities students are completely free to omit the lecture part of this and simply do self-study worksheets and seminars and skip the lectures (and I would support them in that freedom). However, I believe that the lectures are, in fact, useful and the tests you cite do not provide evidence about that, because the students then have the lecture notes to refer to to guide their self study. If the study compared students who went to lectures, did worksheets, revised for exams and sat the exams versus students who did self-study for an hour (or read through the lecture notes) then that would be relevant data.

So the lecture alone doesn't put the material in the students brain... they actually have to look at the notes and read the texts. This is, in itself a hugely valuable skill since it's what a lot of people will need to do in their later working life. I would estimate a decent student should be doing 4 or 5 hours of self-study for every hour I lecture. (that said, by that measure I was a poor student).

Since I have a pile of exams sat on my desk I shall see pretty much exactly how well this works.

[andrewducker]

If the study compared students who went to lectures, did worksheets, revised for exams and sat the exams versus students who did self-study for an hour (or read through the lecture notes) then that would be relevant data.

Absolutely. In the meantime I'll continue with my hypothesis that lectures are less useful than reading an introductory book, which would contain material _like_ lecture notes, only more accurate and better written.

[steer.livejournal.com]

which would contain material _like_ lecture notes, only more accurate and better written.

Well, I'm not 100% in agreement with better written since the person who wrote the book is often the person who wrote the lecture notes (and, in fact, lecture notes sometimes become books)... however, it would take a skilled student to pick the material for the course from the book... so, to get the equivalent material for my most recent lecture notes the students would have to read selected chapters from five different books -- I mark which books those are for them should they wish to do that -- they will have a much broader understanding of the subject but it will take them a lot longer -- plus I preface the lecture notes with the introductory material which the text books often miss out or, say, spread over the first five chapters.

I hope it's not arrogance, but to pass my choice a student would have a very very hard time getting that material from text books as I distilled the course to just include those parts it was necessary for me to teach from the many text books I've read.

There are advantages to the text book -- it provides a broader understanding and may have been proof read more -- although now my lecture notes for this course are on their 4th or 5th year through they have the majority of the major bugs worked out.

[andrewducker]

You would seem to be a better lecturer than many of the ones I had teaching me.

Although at least some of the better ones provided their own printed lecture notes, as they realised that spending a lecture half-listening as you frantically scribble is a particularly stupid waste of time.

[steer.livejournal.com]

Yes, these days, all course notes for courses in my dept are on the departmental virtual learning environment. While it may seem non-eco friendly, they are also printed and given to students before the course start -- otherwise students tend to spend the time writing. I also have copies on my own website.

http://www.richardclegg.org/lectures/

http://www.richardclegg.org/ccourse/

(The C programming course stuff is pretty old now -- would not teach that these days).

I also do separate slides and lecture notes, some lecturers just send out copies of the slides but I think that means either (a) lots of explanatory text on the slides or (b) slides which make no sense out of context.

Also UCL puts all our lectures on video (to enjoy and enjoy again) so students can recap the main points -- though I doubt that many do.

I know it's easy to bash a "lecture" as a way to deliver information but actually, listening to a lecture is a skill like any other. As I often attend conferences and need to be able to pick up information from a 40 minute verbal presentation with slides, it's a good skill to have.

[andrewducker]

Yes, these days, all course notes for courses in my dept are on the departmental virtual learning environment.

I think I'd like to use a time machine to take my degree now. I think I'd be a _lot_ less frustrated.

[steer.livejournal.com]

I know what you mean... these days departments very much frown on "chalk and talk" as a learning method. There are people who regard the lecture as "obsolete" but I'm not one.

[andrewducker]

Also, I do think that you can extract information from lectures. But I can do it about five times as fast from the same contents written down, and go back to bits to re-read them if I get stuck.

I guess I just don't see the advantage of someone reading some text to me over me reading it myself.

[steer.livejournal.com]

In my case when I do the lecture I'm also taking questions from students, explaining parts where they do get stuck, asking them if they understand where we are at so far... posing questions for thinking about in the break and giving demos of software.

Students have an opportunity to privately discuss with me at the end as well.

However, for the top-end students, missing the lecture and reading the notes might well be 90% as good. That said, I've regularly encountered high-flying students who miss out on learning because they are good enough to get through without reading lecture notes and sometimes have missing techniques in their armoury -- so I've occasionally taught students who were better coders than I am and sometimes regretted that they felt there was little they could learn by going at the pace of the majority.

I'm perfectly happy for students to skip the lecture and read the notes but unfortunately my institution is not and attendance is compulsory and checked with a register -- I wish it weren't.

[khbrown.livejournal.com]

Did the students who were better coders also have a better idea of the underlying theory and algorithms?

[steer.livejournal.com]

In my experience there was not a strong correlation between underlying theoretical mathematical ability and coding. I was teaching mathematics undergraduates and post graduates so almost everyone had reasonable maths skills. However, it was often the case that outstanding coders were poor mathematicians and poor mathematicians were outstanding coders.

In general I think a good coder only needs to know "this is the efficient algorithm" to code it not really to understand fully how the proof of the order of execution time follows.

[andrewducker]

Most coding, so far as I can tell, is holding together complex patterns in ones head, and building structures out of them while understanding the interactions between them. This is usually better correlated with pattern matching and chunking ability than maths, in my experience.

[steer.livejournal.com]

Hmm... I think the "chunking" ability is critical -- being able to split a task down into logical sub units -- but it's hard to know if other people go about coding in the same way I do.

[andrewducker]

Well, I work with Objects, so it's easy for me to think in terms of those chunks. I suspect that people who work functionally think differently. When I worked in Prolog I definitely thought in a different way after a few weeks with my head in the code. Oh, and when working with highly recursive code I think differently after a while too.

[channelpenguin.livejournal.com]

I follwed that stragegy at uni - if the lecuruer had written a book on the semester's topic ( fairly common!) I just read it (I am a fast readr so that took a few days max, usually) and never bothered with the lectures. I did the assignments, labs and some tutorials - though that's what I should have managed to attend more of as that's the really useful bit!

[naath.livejournal.com]

Lectures taught me how to show up somewhere at 9am ready to do work if they taught me nothing else at all. And I don't think I learned nothing at all in lectures, although I did need to go over the notes, do the exercises, etc. as well.