Delicious LiveJournal Links for 9-5-2009

[andrewducker]

• Maths no better than in 1970s
  3000 children sit an exam from 1976, get pretty much the same results that people got in 1976. This despite massive result inflation since then...
  (tags: education mathematics UK)
• Damien Hirst in vicious feud with teenage artist over a box of pencils
  A box of pencils worth, apparently, £500,000
  (tags: art money UK crime)
• Mixing Solar with Coal to Cut Costs
  Now this is a smart idea!
  (tags: solar coal energy ecology power electricity)
• 50 things that are being killed by the internet
  8 - Telephone directories (can't remember the last time I used one) 22 - Enforceable copyright 44 - Trust in Nigerian businessmen
  (tags: internet culture technology society)
• Dog Hair is Defined by Just Three Genes!
  I was expecting...more
  (tags: dogs hair genetics)
• Research into whether women are more tentative than men
  "I found that women are more tentative than men sometimes, and men are more tentative than women sometimes," Palomares said. "It depends on the topic and whether you're communicating with someone of the same gender. Gender differences in language are not innate; theyâ™re fickle.
  (tags: language gender psychology social stereotypes)
• German boffins invent steel Velcro
  Offers adhesive strength of up to 35 tonnes per square meter when tensile force is applied parallel to the fastener surface. When it is applied perpendicular to the fastener surface, Metaklett can still withstand a force of seven tonnes per square meter. Moreover, like a standard Velcro fastener on a childâ™s shoe, it can be opened and closed again without the help of any tools.
  (tags: velcro technology)
• Arctic warmest in 2,000 years
  (tags: climate environment science)
• Will it Rain Today?
  Rain radar for the UK
  (tags: weather UK maps radar rainfall forecast)

Comments

Teaching to the test bollocks

[endless-psych.livejournal.com]

Article contains bugbear number 2388960: teaching to the test is not a problem with the teaching it's a problem with the test!!!

Saw one of the researchers on the news talking about this didn't sound like they actually made them sit a 1976 exam.

Although given that the kids are being taught a different syllabus now that might just account for the disparity in results...

Also the guy seems to have an issue with inclusion in education:-
He also said mainstream schools today had a higher proportion of lower-achieving pupils, whereas in the 1970s many of these pupils would have been in special schools.
Which isn't the full picture. I blogged on that not too long ago actually:- http://andyourelectronmicroscope.wordpress.com/2009/08/04/social-inclusion-of-children-with-learning-disabilities-in-education/

That and this appears to be yet another science story based upon a conference presentation!!! Jeezo.

Re: Teaching to the test bollocks

[andrewducker]

Article contains bugbear number 2388960: teaching to the test is not a problem with the teaching it's a problem with the test!!!

I disagree. Unless your test consists of a 6 hour chat about the subject, it cannot contain all the details and intricacies of understanding necessary to properly understand a subject. The vast majority of tests are simply based around regurgitation of facts, which means that the teachers simply teach those facts necessary to pass the test.

Teachers should, IMHO, be doing vastly more than that.

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

You can feel free to disagree. Doesn't change the fact that there is still a problem with the test if it can be reduced to a few "regurgitated facts". The test in such a case is still flawed.

What thanks do teachers get for teaching outside the syllabus? Fuck all.

Re: Teaching to the test bollocks

[andrewducker]

The test in such a case is still flawed.
And this is true, in my experience of _all_ tests. So unless you have some new kind of test which is different to all of them then any future tests will be flawed. Which means that teaching to them will be a bad idea if we want people who actually understand and have an interest in a subject.

What thanks do teachers get for teaching outside the syllabus? Fuck all.
Exactly - and that's the problem. People want to measure success, and all measurements cause skew once people realise they are being measured, because they concentrate only on the measured outcome, and not on the bigger picture.

It's a constant problem in businesses that reward people on results too. A lot of businesses rewarded directors based on share price, and then were amazed when directors did stupid things that temporarily maximised share prices to the long-term detriment of the company.

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

if we want people who actually understand and have an interest in a subject.

Do we?

I rather think that people who actually understand and have an interest in a subject will rather find out more themselves on their own time.

Education is more about building a basic skill set of literacy/numeracy and other associated skills then it is about producing new intelligensia.

Re: Teaching to the test bollocks

[andrewducker]

Education is more about building a basic skill set of literacy/numeracy and other associated skills then it is about producing new intelligensia.

Up to the age of 13 maybe. After that point it's pointless if that's all it's doing.

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

Thereafter its about producing people to go out and be productive ;)

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

What is the bigger picture in terms of education btw?

If you can't measure it all you have is a subjective interpretation based on your subjective value of what education should be about. How much money do we throw at that? How do we know if it's effective? How do we know if anyone is learning anything or engaged or interested?

Re: Teaching to the test bollocks

[andrewducker]

Oh, absolutely. This is a constant problem all over the place. Same with the NHS, and also in businesses.

For instance, measuring the productivity of programmers is notoriously hard. If you measure them by lines of code then they churn out long complex code that's hard to maintain. If you do it by bugs fixed then bugs get put in just to be fixed again. If you do it by number of modules then code gets split into many more modules than are necessary. If you don't measure it at all then how do you know if they're doing a good job?

The answer, generally, is to get lots of feedback from peers, trust that their managers can spot a good coder from a bad one, and aren't crap/corrupt themselves - and that if they are, this will become obvious in other ways. But people that like metrics (i.e. high level managers) don't like this, because it can be years before problems become apparent, and it looks like they aren't in control.

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

Another thought: Not teaching to the test and teaching the broader subject probably would have the unintended consequence of making education become more elite.

As children with access to more educational resources out of school will progress faster and further then those without. So children from middle and upper class families will likely ask more questions, demand more of the teachers time in exploring obscure (or otherwise) areas of interest. Which will monopolise the teachers time in dealing with such things and dealing with the engaged and interested children. Whereas those who aren't interested may fall by the wayside. A frightening prospect in Maths and English.

Thus education could concievably proceed upon a continuum of wealth. With the poorer in society suffering. Perhaps creating a self perpetuating (more so then is currently evident that is) circle of generational poverty.

Re: Teaching to the test bollocks

[andrewducker]

This is pretty much the situation as it stands. The problem is teaching people of strikingly different abilities together - with the teacher having to teach all of them at the same speed. I am convinced that most of the people I know who think they are bad at maths were simply left behind at an early age.

Not having to work towards a standard test - with the teachers job being to focus on the people who would get the highest marks (and thus push the school up the league tables) would improve this.

Re: Teaching to the test bollocks

[endless-psych.livejournal.com]

Not having a standard test would excacerbate and disadvantage those who are already disadvantaged further. As you would have a situation akin to people being able to buy progress and a poverty of resources would unfairly punish those on lower incomes.

People with striking different abilities aren't taught together from age 13 or indeed for the key subjects (maths and english) before that. There is a fair degree of seperation in what people learn, depending on ability at Primary level. We're a fair bit past the days of Victorian rote learning but at the same time, nessecarily due to funding, away from the ideal situation of fully individualised learning.

Teachers focus on the children they think are brightest anyway, research has shown this time and time again - before and after the advent of standardised tests.

I think that the idea that teachers would focus on the small number of children who will get high grades because of standard tests is logically flawed. If anything it would force teachers to balance the costs (in time and attention) with the benefits (of raising the grades) of which children they focus on. thus teh worst and best kids should get the least focus and the kids who are "middling" will get the most attention. As it invovles the least cost for the most benefit.

Re: Teaching to the test bollocks

[andrewducker]

I think that the idea that teachers would focus on the small number of children who will get high grades because of standard tests is logically flawed.

School grade reporting is frequently on the people who got As. As it is you get schools forbidding pupils from taking exams they might not get good grades in, in case they bring the average down. I've seen examples of both of these things reported in the past - and the latter has a few instances that I know of amongst my friends.

[anton-p-nym.livejournal.com]

Can we just take both of the "artists" in the pencil fracas and shoot them? Please? This is beyond petty, to the point that I'm no longer certain that "up against the wall, [censored]" is inappropriate.

-- Steve does think, on a tangent, that "Steel Velcro" is so the name of his next band. (And it's a really cool concept, if it proves out.)

[andrewducker]

Yes. "You're a pair of six year-olds, fuck off!" is probably the best response.

And I want some Steel Velcro for hanging shelves!

[johncoxon.livejournal.com]

I'm not sure I can be bothered to read the massive argument entitled "Teaching to the test bollocks" but I agree that a part of the disparity could just be that the curriculum is different now. For instance, I learned matrices in Further Mathematics at AS-level, whereas, according to my old maths teacher, they used to be taught before GCSEs. That's not necessarily a sign that the teaching is dumbing down, just that the priorities of the curriculum are shifting (matrices are not terribly useful to anyone who isn't planning to go into maths or physics-related study, so why bother teaching them before students have a chance to start specialising?).

It's impossible to get a class of 3,000 1976 schoolchildren to sit a 2009 exam, and IMO no useful conclusions can really be drawn if you can't do the test both ways.

[andrewducker]

Yeah - there will undoubtedly be some things that have moved back and forth in importance. Logs, for instance, would have been vital before there were calculators, but are nowadays not so vital.

I suspect that they were able to compare on a bunch of different areas that haven't changed that much - and I'm not too troubled by that. I don't think that teaching is dumbed down nowadays - I suspect that the people doing GCSEs nowadays are learning about as much as the ones who did O levels then - but I suspect the grade inflation is a real thing. Hard to prove though.

[marrog.livejournal.com]

Ah, but then you come up against the wood glue problem - the bond will still only be as strong as the bond between the 'velcro' and the object its attached to. It's pretty easy to break apart two piece of wood attached with PVA glue - the glu holds, but it rips off the surface of the wood. What do you use to hold the velcro to the building blocks?

[andrewducker]

MOAR VELCRO!

[andrewducker]

Presumably they can use rivets or massive bolts. You'd only use Velcro for when you had interchangeable bits to connect to it.

[marrog.livejournal.com]

Don't you have to be able to bend it back to peel it? How do you detach something if you can't peel the velcro because it's attached, flat, to the block? Wouldn't you have to unbolt it somehow first? And if so, why not just bolt the block onto the other block in the first place?

[andrewducker]

Good point - I was only thinking of the point you were attaching it to.
Hmmmm.

[marrog.livejournal.com]

Hm, have thought of some answers to my own question, nevermind.

[andrewducker]

Share!

[marrog.livejournal.com]

You would have to use the velcro like a sticky belt and stick joists holding things together - the way you would with single sided sticky-tape rather than double sided. Which will of course mean they can only carry the 7 tonnes and not the 35, because the pull will usually be in the wrong direction ie along the connection rather than away from it.