Interesting Links for 26-02-2015

[andrewducker]

• Some student-professor pairings lead to "unusually effective teaching" (and it's possible to predict which ones)
  (tags: learning teaching )
• EU referendum: Support for Britain's EU membership reaches all-time high
  (tags: Europe )
• Clampdown on cold call companies unveiled by government
  (tags: phones UK government spam )
• How to tell that you're not a

Comments

[alitheapipkin]

To be fair to the 'How to know you're really a writer...' author, she does say that the last point - writers write - is the only one you have to agree with... But yeah. I mostly find people talking about 'real' anything when they don't mean as opposed to literally fake, are generally saying things that annoy me.

[alitheapipkin]

This comment really doesn't work as well now I've realised it's the person commenting on how daft it is who used the 'real writers' phrase... Oh well.

[a_pawson]

Playing Space Invaders is one thing, but if anyone lets that A.I. loose on Noughts & Crosses and Global Thermonuclear War we're in trouble.

[harvey_rrit]

A journal for a profession which in its entirety has never produced a cure since it was established has banned tests to learn the chance of getting the same result or better from doing nothing.

Thank you. This is something I could never have made up.

It'd be a wonder why they didn't think of it before, if it weren't such a wonder that they thought of it at all.

[a_pawson]

You know there is a Journal of Homeopathy right? In fact there are a whole bunch of them.

[harvey_rrit]

That doesn't make me feel better, for some reason.

[apostle_of_eris]

Wasn't the "three thirds" study from the 1950's?
Someone who couldn't get real patients to work on compared people who got treatment with people who didn't. One third got better, one third got worse, and one third stayed the same - for both groups.

[amaebi]

I, by contest, didn't mind at all the banning of [reporting] significance test results. In my experience in economics, they really are reported in a thoughtless, knee-jerk way, with their meaning seldom thought about and so far as I can tell, often unknown to researchers and readers.

[andrewducker]

Yup, much as the article says - they're a really crude metric that's vastly overused without understanding.

[pennski]

I was sad that the bullshit meter did not recognise "pre-align" or "low-hanging fruit".
Ah well, we all have our buzzwords.

[andrewducker]

Make a recommendation!