Thursday 27 August 2009

Evaluations

I'd manage to totally forget the abolition of maths coursework, but I feel no remorse whatsoever at its passing; it was a monumentally hideous and stressful task to corral children into getting it done in even the most cack-handed fashion, and the possibility of plagiarism made the whole thing pretty pointless (this is accentuated by how much I hated maths coursework as a child, though since I most definitely fall into the "cram at last minute" crowd, I guess that just makes me a stereotypical male). If you're going to make teachers (and those pupils who actually do give a damn about their progress) go blind with stress for a couple of fortnights a year, you should at least be sure you're getting something out of it.

The most interesting thing here (for me at least) is the suggestion that boys and girls should be assessed differently (for the sake of argument, we'll have to ignore the plagiarism aspect for now, since "should girls be examined in a way that makes it easier to cheat?" might throw off the debate otherwise). Off the top of my head, it would take some time to adjust to the new system, causing a dip in marks for a while, and I don't think it would do much for relieving teacher stress. There may be some philosophical/ethical issue to this as well, though I haven't yet come up with anything specific. I think the biggest problem, though, would be entirely practical. Coursework can only cover so many topics (one fortnight of geometry and another on data analysis, in the case of my teaching experience), so absent far more tasks, girls would still be required to sit a final test, simply with a lower weighting attached to it. The immediate question then becomes: what are all the lads doing during those four weeks? Are they doing the same task but for no credit? Are we doubling the number of maths teachers (or, more sneakily, swapping kids around to double the ability range of each class, which would bring its own problems)?

Or is the plan to modularise this stuff? Can a pupil choose the weight of their own exams, given the results of the coursework (again, assuming we could eliminate the cheating issue)? Since we already have modular courses in place, in which one can resit indefinitely until you run out of money or will (or have to get a job), I'm not sure it would be too much of a stretch. This would help avoid having to run parallel classes (though the temptation for students to blow off coursework because "they'll do it on the night" could well be pretty strong), and allow children to choose an assessment method that suited them personally, rather than attempting to make gender-specific changes across the board.

It's an interesting idea, at least.

While we're on the subject of interesting, I note that English results have had a wobble. I await the upcoming flood of people confidently stating that this is obvious proof that English exams are getting harder. I mean, that's the only possible conclusion, right?

10 comments:

BigHead said...

Nah, obviously the English results got worse because they underestimated how much they needed to dumb things down this year :)

For the coursework vs exams question, it seems like the skills, abilities, knowledge and attitudes required for success in these discliplines are somewhat different. The weighting between the two should be determined by considering the relative importance of the differences in the given subject.

Doing maths coursework is very different from doing a maths exam, and if it's thought that a maths GCSE should include aptitude for the sorts of things required for coursework, then it should be included, the weighting kept the same for everyone, and if boys aren't very good at it then they'll just have to live with being not as good at maths GCSE. Similarly if it's thought that coursework skills aren't really what maths GCSE is about then they can be weighted pretty low or removed altogether, and if this inconveniences girls then so be it.

On the other hand having a couple of different weightings that can be chosen and end up giving a different name of qualification doesn't seem too unreasonable in principle. What it would mean in practice is not for me to speculate on.

I think that whatever, the choice should be individual as you say. rather than enforced by gender, to avoid victimising people who don't fit their gender's pattern.

Chemie said...

'There may be some philosophical/ethical issue to this as well, though I haven't yet come up with anything specific.'

Treating boys and girls as fundamentally different intellectually is wrong? And doing so only leads to greater diversity? If the level of the mathematical ability expected of a 16 year old resides in 90% exam skills then 'typical girls' will just have to be bad at maths. If coursework skills are an important part of maths development, then it sucks to be a 'typical boy'. Modular systems also favour the 'more mature' girls.

It would be far more beneficial to look at *why* the genders supposedly handle the skills differently and address the root cause. (And I'm one of the girls who kicked arse with the last minute exam cramming)

SpaceSquid said...

"Treating boys and girls as fundamentally different intellectually is wrong?"

My gut instinct is to agree, but this needs more than just "it's wrong", which is why I originally said that I couldn't come up with anything specific. I'm guessing you put "intellectually" in your comment since we already seem entirely happy with treating the genders as fundamentally different physically. Is it a case of not wanting to do anything about the intellectual issue without far more conclusive evidence that such an intellectual difference exists (which is the obvious difference between how to judge people in maths lessons vs how to judge them in games lessons)? I would certainly have a lot of sympathy for that position, if nothing else.

"And doing so only leads to greater diversity? "

That does seem likely, and again a priori it's an idea that makes me uncomfortable. If we assume the disparity is real, though, is that possibility really worse than knowing you are hobbling one half of the school population in order to maintain the illusion of a balance that doesn't exist? It's at least arguable that reducing diversity (or fighting to prevent it increasing)is only a worthwhile goal when the diversity isn't real, or when the diversity is real, but irrelevant to the measure being applied.

I am trying very hard here to walk a line between wanting to acknowledge people's differences and not sliding into the kind of bell-curve bullshit that makes me so mad whenever it rears its head. With this in mind, then, I am happy to recognise the possibility that even if the difference suggested proved to be real and problematical, it could be concluded that it was a necessary evil to prevent it being used as an excuse for gender discrimination down the road. Of course, if a smaller proportion of female applicants get onto university maths courses because the current exam system favours males, it's not like the case for the current system employing de facto gender discrimination wouldn't have any ammunition either.

Chemie said...

'Is it a case of not wanting to do anything about the intellectual issue without far more conclusive evidence that such an intellectual difference exists (which is the obvious difference between how to judge people in maths lessons vs how to judge them in games lessons)? I would certainly have a lot of sympathy for that position, if nothing else.'

I don't think there is sufficient evidence to show that there is an inherent difference in the gender's brains. All of these brain wiring theories use the concept of a scale. So they can't be used as evidence to support gender discrimination. I absolutely hate any and all attempts to generalise about men and women (beyond the obvious physical stuff). We are individuals and we should be treated as such. We don't have the evidence to make strong statements about how the genders think and what we do have leads us to believe a huge chunk is environmental and rest places us on a sliding scale.

'If we assume the disparity is real, though, is that possibility really worse than knowing you are hobbling one half of the school population in order to maintain the illusion of a balance that doesn't exist?'

If the disparity is real, then what matters is the nature of the goal. You can do maths or you can't. You don't change the maths or the goal. You can change the teaching method if you want (and hopefully fairly) but not the assessment. You still have to meet what are accepted as reasonable necessary requirements of 'Maths GCSE'. There is a good example associated with the military. Which (used to?) say women couldn't do certain jobs, based on the fact that they weren't strong enough. The alternative idea is to simply set the requirements for a job appropriately and accept people who can meet them. It should be the same with maths. Set the requirements as expected by the maths community and employers (ability to use quadratics, handle probabilities, handle data, follow logic arguments or whatever) and people pass who meet the requirements in the assessment method that best illustrates them.

SpaceSquid said...

Another two-part reply, I'm afraid.

"I don't think there is sufficient evidence to show that there is an inherent difference in the gender's brains. All of these brain wiring theories use the concept of a scale. So they can't be used as evidence to support gender discrimination. "

So it is just a matter of whether or not the evidence is there. That's entirely fair enough.

"I absolutely hate any and all attempts to generalise about men and women (beyond the obvious physical stuff). We are individuals and we should be treated as such."

Of course, the current system in no way treats us like individuals either (excepting streaming by ability, which no-one is talking about doing away with). In fact, if it could be shown that this hypothetical difference between male and female populations does in fact exist, then a gender-differentiated teaching system would treat people more like individuals than the current one does. Certainly, once the disparity is proved, the fact that outliers would certainly exist under the new system doesn't change the fact that more outliers would exist under the old.

"You can change the teaching method if you want (and hopefully fairly) but not the assessment. You still have to meet what are accepted as reasonable necessary requirements of 'Maths GCSE'."

I agree entirely that you need to show you've met the same requirements, but I don't see how that in any way prevents different assessment. The goal of a maths GCSE is not to demonstrate how well someone remembers mathematical facts without being able to reference them, any more than it is to see how good they are at a mathematical investigation. It is this disparity between how close a pupil is to the goal and how close the assessment methods used claim they are to the goal that is being addressed here. I agree that, aside from the possibility of individuals choosing their own weightings, the best idea would be to try and tweak the system so as to lie at the exact mid-point of any disparity, but the formulation "Boys are better at X, girls at Y" can be easily re-jigged into "Boys are more accurately assessed by X, girls by Y". If exams really are a bad way to assess girls, then "tough luck" isn't really a very compelling response.

SpaceSquid said...

"There is a good example associated with the military. Which (used to?) say women couldn't do certain jobs, based on the fact that they weren't strong enough."

I don't think that's a good example at all, for several reasons. Firstly, a Maths GCSE is designed to measure a person's general mathematical skill (which, to be fair, you address later). A military strength test is designed to measure a person's aptitude for the military. The added specifity makes the possibility of inaccurate assessment less likely. The second reason may reveal my ignorance of the way these tests are performed, but I would be surprised to learn that the military strength test offers as many potential options for the manner of testing (strength not being the same thing as more general fitness, of course), and in addition might well involve a task that is either pased or failed (doesn't it involve carrying a given amount of weight a given distance, or something?), rather than attempting to assess hundreds of mathematical ideas, all of which will be understood or not to vastly varying extents (I still don't really get a lot of mechanics, for example).

In sum, "You're strong enough for the military or you aren't" is a much more convicing statement than "You're can do maths or you can't"

"Set the requirements as expected by the maths community and employers (ability to use quadratics, handle probabilities, handle data, follow logic arguments or whatever) and people pass who meet the requirements in the assessment method that best illustrates them."

Isn't that what the National Curriculum is already? For all the lip-service maths teachers give to the idea of maths being useful in daily life, it's really almost entirely about the job market.

Chemie said...

'If exams really are a bad way to assess girls, then "tough luck" isn't really a very compelling response'.

But if an exam is the best way to test a skill then it really is, 'tough luck -you're a girl'. My point with the military is that it is better to set a list of requirements and see who can pass them, rather than make assumptions about gender-based abilities (strength, sight, calmness under pressure, aggression etc ). Dealing in facts, not gender based supposition makes sense. So that it is better described as 'You can do maths to the standards expected by the exam board, or you can't'. It's the standards and how they are assessed that is important. You assess different skills differently. Use the assessment that tests the skill the most fairly, if that sucks for someone who prefers another method, then it sucks. If two methods are comparable then let the candidate choose. There is also something to be said for collecting all skills together in an end of year assessment (which supposedly favours the slow maturing boys) so that candidates aren't able to learn something, forget it and move on.

It's far more important to ask why girls are supposedly 'more conscientious' than boys and therefore thrive on coursework?

Dan Edmunds said...

So, am I right in thinking that all this is basically saying is that boys are on average better at Maths, but girls are on average more “conscientious” and thus are able to more than make up for this slight difference if the course includes coursework?

I agree with Kimmy, in that surely you just need to determine if coursework is a worthwhile part of the course, if so bring it back, if it’s not don’t. It’s kind of irrelevant if boys do 2% better because of it than girls or vice versa. If you had one course for girls and a different one for boys all that means is that girls would end up being better at one aspect than boys and boys would end up better at the other aspect. Then when it comes to A-Levels and Degrees or anything further down the road, unless you keep the segregation your going to disadvantage one group or the other at that stage. At least by teaching them all the same they get to attack the next stage on a level playing field.

SpaceSquid said...

@ Chemie

"But if an exam is the best way to test a skill then it really is, 'tough luck -you're a girl'."

I wouldn't argue with that, the question we need to consider is whether an exam is the best way to test a skill, and whether it can be the best way to test a skill if it turns out that it consistently undervalues the mathematical skill of half the population. In other words, "best" is not necessarily defined very easily.

"My point with the military is that it is better to set a list of requirements and see who can pass them, rather than make assumptions about gender-based abilities (strength, sight, calmness under pressure, aggression etc )."

And my point is that the wisdom of that suggestion is conditional on how sensibly one can produce said list of requirements. This, again, is why I am unconvinced by the military analogy. The military test wants to know if you can do what the military requires. The maths GCSE attempts to guage your overall mathematics capability. The latter cannot be reduced to "facts" in the way you seem to mean them, because mathematical comprehension is a black box in a way that having sufficent strength to be a soldier is not.

"It's the standards and how they are assessed that is important. You assess different skills differently. Use the assessment that tests the skill the most fairly, if that sucks for someone who prefers another method, then it sucks."

I agree, 100%, but that immediately leads to the $64 000 question: why you think gender-conditioned assessment is less fair. So far all you've really offered is "We should be treated as individuals", despite the self-evident point that while gender-conditioning might not be a very good idea, the very last thing refusing to try it does is increase the degree to which pupils are treated as individuals.

All this touches on a much wider question, which is: to what degree is "equal treatment" the same thing as "fair treatment"? In this case, it seems entirely obvious that if two people can be equally mathematically able but a gender-variable imprecision between true ability and assessed ability gives one child an A and the other a B, then equal treatment has not led to fair treatment. Once the bias has been demonstrated, you can no longer have both of these at the same time, which means you have to choose. Whichever one you do choose, you can sensibly back up, of course, but you seem to be trying to have both at the same time.

"It's far more important to ask why girls are supposedly 'more conscientious' than boys and therefore thrive on coursework?"

It is most certainly a excellent question, but while anecdotally the hypothesis seems to be reasonable, I'm less concerned with whether it's true, and more concerned with whether a bias, once proven, should be acted on, whether or not a specific explanation for that bias is correct or persuasive.

Moreover, the assumption that there exists a "root cause" that can be tackled is just that, an assumption. It may be genuinely be biological. It may also prove to be behavioural, and indicative of the differences in which boys and girls are raised, in which case, while I have sympathy with the idea that we should attempt to educate parents on such matters, I always get a little antsy around the idea that until parents just shut up and listen, we're going to knowingly feed their kids into a biased system. To use another analogy, educating people in the idiocy and dangers of bigotry doesn't mean not dealing in employment quotas. Equal is not the same as fair.

SpaceSquid said...

@ Spielbergo

Hopefully a lot of what you raised has already been covered above, but I just wanted to flag this.

"At least by teaching them all the same they get to attack the next stage on a level playing field"

I think you're confusing differences in assessment with differences in teaching. Of course, the former might well lead to the latter, because of the pressure on teachers to chase marks, which is one of the major practical objections to trying this. It is still important to note however that we are absolutely not talking about, though, is offering two different curriculums. The question is whether girls and boys differ in the way they can be most accurately evaluated on the same taught material.