not to diss your students but most of uni these days is cramming stuff into your head a week before the exam and then hoping you can jackson pollock your mental diarrhea into something coherent. There's only a handful of them spending any appreciable amount of time in the library.
LOL love that description of the learning process
Like @klemming said, we largely switched to coursework
But it takes so long to mark - I miss exams!
Also, I am happy to report our students spend plenty of time in the library... Drinking and partying!
(really - turns out people were using the group study spaces for parties before the lockdown hit X)
0
Options
HerrCronIt that wickedly supports taxationRegistered Userregular
Trying to cram for a week (or two if you had your shit together) was definitely the way of things when i was in uni which was ....twenty years ago. Fuck me where does the time go?
I don't know how far you'd have to go back for that not to be true.
Students - far more traditionalist than you'd expect.
There is going to be a week of heart wrenching headlines as straight A students from deprived schools are marked down.
The SNP are fucking idiots for not seeing this coming.
I can sum it up too. Most deprived areas generally have greater number of BAME cohorts. Once it's broken down by ethnicity suddenly the SNP and the SQA are going to look 1000x worse than they already do.
The biggest fear is that when it's time for England's results I can see an even bigger fuckup coming.
I mean AQA are already sending letters to schools offering students unhappy with results or wanting better grades to take a fast track GCSE and A level exam throughout the autumn term. You know, year 11 and 13 students who have already left school/ college, to bring them back into already over capacity schools to take a 1 unit only exam. The same students who will be starting College or university in September..
I'm getting serious "That Yellow Bastard" vibes from it
I was thinking more "evil vizier" and/or "that Wormtongue energy".
If he doesn't do a steeped hands and say "Everything is proceeding as I have planned" in every appearance, that'll be a missed opportunity.
followed by someone pointing out a major flaw in his plan and him doubling down on it.
"My liege ... the satellites we purchased ... they're the wrong specs!"
"Everything. As. Planned.", he says, with a smile that's intended to intimidate but screams "I just shat myself".
I'm getting serious "That Yellow Bastard" vibes from it
I was thinking more "evil vizier" and/or "that Wormtongue energy".
If he doesn't do a steeped hands and say "Everything is proceeding as I have planned" in every appearance, that'll be a missed opportunity.
followed by someone pointing out a major flaw in his plan and him doubling down on it.
"My liege ... the satellites we purchased ... they're the wrong specs!"
"Everything. As. Planned.", he says, with a smile that's intended to intimidate but screams "I just shat myself".
Fortunately, this is the default look of Spitting Image puppets.
I'm getting serious "That Yellow Bastard" vibes from it
I was thinking more "evil vizier" and/or "that Wormtongue energy".
If he doesn't do a steeped hands and say "Everything is proceeding as I have planned" in every appearance, that'll be a missed opportunity.
followed by someone pointing out a major flaw in his plan and him doubling down on it.
"My liege ... the satellites we purchased ... they're the wrong specs!"
"Everything. As. Planned.", he says, with a smile that's intended to intimidate but screams "I just shat myself".
The only idea I've seen that makes any sense is for people intending to go to university to sit an entrance exam there and throw the school grades in the bin. Of course that would mean universities have to pull the funding and infrastructure to hold many thousands of examinations out their arses. Not to mention they'll face the same difficulties with distancing that caused schools to cancel their exams in the first place.
All that said it still makes more sense than "we give you a random grade based on how affluent your area is and if you don't like it tough titty".
The only idea I've seen that makes any sense is for people intending to go to university to sit an entrance exam there and throw the school grades in the bin. Of course that would mean universities have to pull the funding and infrastructure to hold many thousands of examinations out their arses. Not to mention they'll face the same difficulties with distancing that caused schools to cancel their exams in the first place.
All that said it still makes more sense than "we give you a random grade based on how affluent your area is and if you don't like it tough titty".
I can't speak for any other universities, but right now the one I work at absolutely doesn't have the time or money to arrange entrance examinations. That's not a trivial task and we're already working at 150% in order to get (mostly?) online for Sept/October.
Or just add a foundation course. Infrastructures there already in many cases and that's probably split a class if it was heavily suggested but not compulsory.
The only idea I've seen that makes any sense is for people intending to go to university to sit an entrance exam there and throw the school grades in the bin. Of course that would mean universities have to pull the funding and infrastructure to hold many thousands of examinations out their arses. Not to mention they'll face the same difficulties with distancing that caused schools to cancel their exams in the first place.
All that said it still makes more sense than "we give you a random grade based on how affluent your area is and if you don't like it tough titty".
Should have you know, just trusted the teachers and the grades given by them as professionals.
Schools funding in Scotland is controlled by local authorities, not central government
There isn't a system that ties funding to exam results. To the extent that there is a variable funding model it's tied to indexes of deprivation
There's a breakdown in the sqa announcement of adjustments by SIMD (Scottish index of multiple deprivation) which explicitly shows that the majority of adjustments have been applied to the most deprived students
(First tweet I found with a decent crop, but the account is apparently that of an education researcher)
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
Schools funding in Scotland is controlled by local authorities, not central government
There isn't a system that ties funding to exam results. To the extent that there is a variable funding model it's tied to indexes of deprivation
There's a breakdown in the sqa announcement of adjustments by SIMD (Scottish index of multiple deprivation) which explicitly shows that the majority of adjustments have been applied to the most deprived students
(First tweet I found with a decent crop, but the account is apparently that of an education researcher)
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
"If your school had no historical data, your teachers' estimates were just accepted. Astonishing."
The scandal is A. using estimates AT ALL, B. knocking down poor students so much worse. And you can see in the data posted that it wasn't anywhere near the same for everyone!
Worst off students:
85.1 to 69.9. 15.2 points, 82% of previous value
Best off students:
91.5 to 84.6. 6.9 points, 92% of previous value
The worst off students were hit much harder. This wasn't "oh everyone dropped by the same amount"
and of course grading students based on historical averages is just..what? You're telling me that a student that was doing perfectly before is going to fail now, for Reasons? Come on. You might as well draw grades out of a hat.
It's only vaguely related, but I've always loved the implicit assumption every year that whenever results rates go up it's because the exams are easier and when they go down it's because the students are stupider.
It's simply not acceptable for students to get smarter, or teachers to get better at teaching a subject.
Nobody remembers the singer. The song remains.
+12
Options
daveNYCWhy universe hate Waspinator?Registered Userregular
At the very basest level I think the message the regrading conveys is, "Know your place."
It's immoral and everyone involved should be fired out of a cannon into the sun.
Shut up, Mr. Burton! You were not brought upon this world to get it!
The glib version of the response to the "teachers were too optimistic" argument is essentially:
Why are we taking past exam results as reliable indicators of ability, when teachers and others have been screaming for years that they aren't?
Call me cynical but I've believed for years (based on the opinions of teachers who know about these things) that exams are a shit way of judging ability. Rich people love them because they're another system they've learned to subvert for their own benefit.
My parents (who aren't even rich) got me an English tutor before I took my highers. I ended up getting an A not because I was smart or talented or thoroughly understood the subject. I got an A because my folks hired a guy who essentially gave me the exam strategy guide on a plate. I went in there with a template in my head of what a grader will slap an A on. I sat the exams. I passed. It was forgotten forever a month later.
Schools funding in Scotland is controlled by local authorities, not central government
There isn't a system that ties funding to exam results. To the extent that there is a variable funding model it's tied to indexes of deprivation
There's a breakdown in the sqa announcement of adjustments by SIMD (Scottish index of multiple deprivation) which explicitly shows that the majority of adjustments have been applied to the most deprived students
(First tweet I found with a decent crop, but the account is apparently that of an education researcher)
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
"If your school had no historical data, your teachers' estimates were just accepted. Astonishing."
The scandal is A. using estimates AT ALL, B. knocking down poor students so much worse. And you can see in the data posted that it wasn't anywhere near the same for everyone!
Worst off students:
85.1 to 69.9. 15.2 points, 82% of previous value
Best off students:
91.5 to 84.6. 6.9 points, 92% of previous value
The worst off students were hit much harder. This wasn't "oh everyone dropped by the same amount"
and of course grading students based on historical averages is just..what? You're telling me that a student that was doing perfectly before is going to fail now, for Reasons? Come on. You might as well draw grades out of a hat.
This should highlight that MSP grilled the SQA overeats process back in May seeking reassurances that the SQA wouldn't just mark down due to the affluent nature of the schools area or its historical results. The SQA failed to address any of these.
The Scottish Greens' education spokesperson, Ross Greer, tweeted: "I'm now more distressed than ever by the system the SQA has set out to replace this year's exams. Profoundly unfair and utterly statistically flawed".
University of Glasgow education researcher Barry Black tweeted: "A chance this morning for the SQA to instil confidence in their plans. Missed. It’s clear now that previous school attainment will be used, and disadvantaged young people will lose out."
The SNP have come out and said that if they had just taken teachers recommended grades then pupils would have performed vastly better than in every other year. Maybe there is some truth to the argument teachers overestimate their pupils? It could certainly be argued there is a slight conflict of interest there since its obviously better for the teachers when their pupils and schools in general are producing better results and unfortunately exam results are the only metric anyone cares about.
Even so, I still can't believe that given a choice between everyone getting slightly better grades than they may have deserved or marking all the poor kids down to balance the stats this is what the SNP/SQA thought was the lesser evil. Sure if everyone got good grades you'd have a lot of right wing papers screeching about "muh meritocracy*", but you wouldn't have a full on uprising by parents and teachers in addition to the negative press. I don't know, does anyone think this system has any integrity at this point?
*Read: What's the point in rigging the game if there are no losers to make me look good?
It's only vaguely related, but I've always loved the implicit assumption every year that whenever results rates go up it's because the exams are easier and when they go down it's because the students are stupider.
It's simply not acceptable for students to get smarter, or teachers to get better at teaching a subject.
Can't agree more!
I've gotten in trouble more than once as a teacher because my students "scored too highly" on an established exam.
What the...?
I'd work my ass off helping them prepare and they worked their asses off studying, taking practice tests, doing group revision sessions etc. I feel like, after students actually succeed and perform highly on a difficult exam, they should be praised for their hard work and not have their marks artificially lowered because "last year's cohort didn't do that well"
So, because of this kind of thing, I'm glad most of my classes are coursework-based now
(even though I know exams can be really useful at times and all that,
Other than having an outlier year in the stats which shows pupils got slightly higher than average scores this year, can anyone reasonably explain the knock on effect of that? Do universities suddenly get swamped and demand outstrips supply? Are there funding implications?
As mentioned earlier, given the two choices of marking pupils up, or marking pupils down, why is it seen to have been the better decision to mark them down? It seems like the negative consequences of that decision are far more impacting and so I'm just utterly confused by the decision here.
The SNP have come out and said that if they had just taken teachers recommended grades then pupils would have performed vastly better than in every other year. Maybe there is some truth to the argument teachers overestimate their pupils? It could certainly be argued there is a slight conflict of interest there since its obviously better for the teachers when their pupils and schools in general are producing better results and unfortunately exam results are the only metric anyone cares about.
Even so, I still can't believe that given a choice between everyone getting slightly better grades than they may have deserved or marking all the poor kids down to balance the stats this is what the SNP/SQA thought was the lesser evil. Sure if everyone got good grades you'd have a lot of right wing papers screeching about "muh meritocracy*", but you wouldn't have a full on uprising by parents and teachers in addition to the negative press. I don't know, does anyone think this system has any integrity at this point?
*Read: What's the point in rigging the game if there are no losers to make me look good?
Well there are other factors that come in from trends with teacher awarding grades. Now I do not doubt that some schools will have teachers that vastly inflate marks for grades, but for the majority will be accurate. What teacher can't predict though are factors such as exam pressure, cramming, aspects of the test that an individual pupil might be stronger at than others, the testing environment in general, etc. All of which can have both negative and positive effects on grades. There will be countless students who are excellent throughout the years but struggle in a test environment and vice versa. Teachers can only go off what evidence they had.
What the SQA have done though is create a literal postcode lottery system with grades.
It's only vaguely related, but I've always loved the implicit assumption every year that whenever results rates go up it's because the exams are easier and when they go down it's because the students are stupider.
It's simply not acceptable for students to get smarter, or teachers to get better at teaching a subject.
Can't agree more!
I've gotten in trouble more than once as a teacher because my students "scored too highly" on an established exam.
What the...?
I'd work my ass off helping them prepare and they worked their asses off studying, taking practice tests, doing group revision sessions etc. I feel like, after students actually succeed and perform highly on a difficult exam, they should be praised for their hard work and not have their marks artificially lowered because "last year's cohort didn't do that well"
So, because of this kind of thing, I'm glad most of my classes are coursework-based now
(even though I know exams can be really useful at times and all that,
Does this rationale make sense to you?
Not to me...
I firmly believe my marking is fair and reliable and high marks are a result of our students' exceptional effort and not artificial grade inflation. Saying that students are only allowed to earn a small percentage of first-class degrees regardless of how well they perform seems crazy to me...
(but I started my teaching career in the US and it's possible I'm not judging the UK higher ed system fairly here - if so, please do let me know and I'll gladly change my views in response to new data ; )
This makes perfect sense.
You just need to add the fact that those in charge are evil, cruel, petty pieces of shit in a poorly constructed human suits into your calculations.
Does this rationale make sense to you?
Not to me...
I firmly believe my marking is fair and reliable and high marks are a result of our students' exceptional effort and not artificial grade inflation. Saying that students are only allowed to earn a small percentage of first-class degrees regardless of how well they perform seems crazy to me...
(but I started my teaching career in the US and it's possible I'm not judging the UK higher ed system fairly here - if so, please do let me know and I'll gladly change my views in response to new data ; )
I can't speak about the details of your situation in particular, but it's basically a combination of trying to avoid degree inflation, and also attempting to maintain a level of intra- and inter-university reliability. A first from university X and university Y should signal roughly the same standard of work, as should a first in module A and module B in the same department.
If you're presenting your students the same (non-MCQ) exam as previous students received, but they're suddenly all receiving firsts, then sure, maybe it could be that they're just an exceptionally strong cohort or your teaching has particularly inspired them, but if they're not performing similarly well on other modules or exams then it's more likely to be the case that you're just marking them too leniently (especially since some of the pedagogical research I'm aware of suggests quite strongly that effort on individual modules will transfer - e.g., you shouldn't see a spike in attainment that only affects a single module if students are just studying harder). The general aim is to have some consistency.
As for whether temporary grade inflation at A level might impact universities... uh, maybe? It could be that some universities give out more conditional offers than they expect to see filled, and if so then they might find their resources stretched when they have more students in Sept/Oct than they're actually prepared for. But the sector's also expecting a drop in students actually wanting to go to university this year, so that might just wind up balancing things out. Possibly.
I think the potential effect on universities is mostly ignorable, and if there's any year where there should have been considerably more leniency than normal given to who gets to pass their exams, it should have been this year.
Schools funding in Scotland is controlled by local authorities, not central government
There isn't a system that ties funding to exam results. To the extent that there is a variable funding model it's tied to indexes of deprivation
There's a breakdown in the sqa announcement of adjustments by SIMD (Scottish index of multiple deprivation) which explicitly shows that the majority of adjustments have been applied to the most deprived students
(First tweet I found with a decent crop, but the account is apparently that of an education researcher)
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
"If your school had no historical data, your teachers' estimates were just accepted. Astonishing."
The scandal is A. using estimates AT ALL, B. knocking down poor students so much worse. And you can see in the data posted that it wasn't anywhere near the same for everyone!
Worst off students:
85.1 to 69.9. 15.2 points, 82% of previous value
Best off students:
91.5 to 84.6. 6.9 points, 92% of previous value
The worst off students were hit much harder. This wasn't "oh everyone dropped by the same amount"
and of course grading students based on historical averages is just..what? You're telling me that a student that was doing perfectly before is going to fail now, for Reasons? Come on. You might as well draw grades out of a hat.
The issue is that the estimated vs historical data shows that the teachers were doing an awful job of predicting whether or not their students would pass exams (thats what this is. A predicted exam result)
So, the teachers saying " This student was doing well and would have gotten a grade A-C" is clearly weak information, whereas the historical data is well clustered and informative.
The fact that they just used the predictions if there was no historical data is clearly absurd, because the historical data shows the predictions are very poor.
I suppose that what this does show is that because the teachers were so bad at predicting, they should have just abandoned the whole exercise as they just didnt have the information they needed to make fair predictions for individuals, only for populations. And individuals are what matters here.
Does this rationale make sense to you?
Not to me...
I firmly believe my marking is fair and reliable and high marks are a result of our students' exceptional effort and not artificial grade inflation. Saying that students are only allowed to earn a small percentage of first-class degrees regardless of how well they perform seems crazy to me...
(but I started my teaching career in the US and it's possible I'm not judging the UK higher ed system fairly here - if so, please do let me know and I'll gladly change my views in response to new data ; )
I can't speak about the details of your situation in particular, but it's basically a combination of trying to avoid degree inflation, and also attempting to maintain a level of intra- and inter-university reliability. A first from university X and university Y should signal roughly the same standard of work, as should a first in module A and module B in the same department.
If you're presenting your students the same (non-MCQ) exam as previous students received, but they're suddenly all receiving firsts, then sure, maybe it could be that they're just an exceptionally strong cohort or your teaching has particularly inspired them, but if they're not performing similarly well on other modules or exams then it's more likely to be the case that you're just marking them too leniently (especially since some of the pedagogical research I'm aware of suggests quite strongly that effort on individual modules will transfer - e.g., you shouldn't see a spike in attainment that only affects a single module if students are just studying harder). The general aim is to have some consistency.
As for whether temporary grade inflation at A level might impact universities... uh, maybe? It could be that some universities give out more conditional offers than they expect to see filled, and if so then they might find their resources stretched when they have more students in Sept/Oct than they're actually prepared for. But the sector's also expecting a drop in students actually wanting to go to university this year, so that might just wind up balancing things out. Possibly.
I think the potential effect on universities is mostly ignorable, and if there's any year where there should have been considerably more leniency than normal given to who gets to pass their exams, it should have been this year.
From what we're hearing from the Unis that are clients at work, most of them seem to be worrying about at the moment is (disregarding for the moment the practicalities of delivering teaching with covid, and international issues):
- oversubscription of conditional offers (most unis do offer more conditional places than they expect to be taken up), however this is always an issue. It's more of an issue if you aren't a prestige university and are less likely to be people's first choice, because you have to hedge against that by assuming a poorer rate of conversion from offer to enrolment.
- the fact that regardless of the grades awarded, this cohort is going to be of uncertain ability and there's an expectation of higher attrition as more people fail to progress. This is a big, big issue because the university loses those fees for the full duration of a (four year, in Scotland) degree
- some English students that have been offered places by Scottish unis are not going to be funded because the English finding bodies capped places at non-English unis after offers had been made
Interestingly, those unis that do contextual admissions will probably fare better, for a couple of reasons:
- contextual admissions assumes that pupils from deprived backgrounds are systematically undergraded, so they apply a calculation that is basically the inverse of what the sqa has done here, and make offers on the basis that a B from a "bad" school is worth more than a B from a "good" school
- those unis that do this well have made it work largely through investment in pastoral care and better mechanisms of support, which means they're better placed to keep students that struggle, and see lower attrition rates generally
- contextual admissions assumes that pupils from deprived backgrounds are systematically undergraded, so they apply a calculation that is basically the inverse of what the sqa has done here, and make offers on the basis that a B from a "bad" school is worth more than a B from a "good" school
- those unis that do this well have made it work largely through investment in pastoral care and better mechanisms of support, which means they're better placed to keep students that struggle, and see lower attrition rates generally
That sounds like an excellent strategy! Tnx for sharing @japan
- contextual admissions assumes that pupils from deprived backgrounds are systematically undergraded, so they apply a calculation that is basically the inverse of what the sqa has done here, and make offers on the basis that a B from a "bad" school is worth more than a B from a "good" school
- those unis that do this well have made it work largely through investment in pastoral care and better mechanisms of support, which means they're better placed to keep students that struggle, and see lower attrition rates generally
That sounds like an excellent strategy! Tnx for sharing japan
There are different approaches and some of them are described on the UCAS website which had some case studies
I mostly know about it from having people that deal with admissions explaining it to me, but it's particularly a thing here (Scotland) because unis can secure additional "widening access" funding if they can show that they can deliver good academic outcomes for deprived groups as defined by the SIMD measures (basically you get more money the more you improve the academic achievement of the students you recruit)
The upshot is that there has been a fair amount of study of how best to do this, and one of the main takeaways is that issue of grading - a moderately achieving student from a poorly performing school is, on average, a more reliable bet for achieving a good degree than a moderately performing student from a high performing school
The tricky bit is that this is often only apparent if you correct for failures to complete a course of study that are non-academic in nature (medical, mental health, family circumstances, funding, etc), which is why you also need to do the pastoral bit and think carefully about how you allocate student support funding such as bursaries
Schools funding in Scotland is controlled by local authorities, not central government
There isn't a system that ties funding to exam results. To the extent that there is a variable funding model it's tied to indexes of deprivation
There's a breakdown in the sqa announcement of adjustments by SIMD (Scottish index of multiple deprivation) which explicitly shows that the majority of adjustments have been applied to the most deprived students
(First tweet I found with a decent crop, but the account is apparently that of an education researcher)
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
"If your school had no historical data, your teachers' estimates were just accepted. Astonishing."
The scandal is A. using estimates AT ALL, B. knocking down poor students so much worse. And you can see in the data posted that it wasn't anywhere near the same for everyone!
Worst off students:
85.1 to 69.9. 15.2 points, 82% of previous value
Best off students:
91.5 to 84.6. 6.9 points, 92% of previous value
The worst off students were hit much harder. This wasn't "oh everyone dropped by the same amount"
and of course grading students based on historical averages is just..what? You're telling me that a student that was doing perfectly before is going to fail now, for Reasons? Come on. You might as well draw grades out of a hat.
The issue is that the estimated vs historical data shows that the teachers were doing an awful job of predicting whether or not their students would pass exams (thats what this is. A predicted exam result)
So, the teachers saying " This student was doing well and would have gotten a grade A-C" is clearly weak information, whereas the historical data is well clustered and informative.
The fact that they just used the predictions if there was no historical data is clearly absurd, because the historical data shows the predictions are very poor.
I suppose that what this does show is that because the teachers were so bad at predicting, they should have just abandoned the whole exercise as they just didnt have the information they needed to make fair predictions for individuals, only for populations. And individuals are what matters here.
Historical data doesn't signify what a current cohort would get. Results from schools fluctuate all the time. Have you considered factors like change in staffing? or Academy takeover? Was the cohort strong expectations this year or weaker than last year? Going on historical data is extremely foolish and has time and time again been discredited for having a poor reflection on a students actual results.
Have you considered the historical changing of grade boundaries to keep the top grades under a particular percentage by authorities like the SQA and JCQ?
Teachers were fine at predicting grades. They always have been.
So it turns out that those mysterious PPE purchases (via Pestfix, Crisp Websites etc) that were rushed through with little oversight were even worse than we thought.
Firstly, none of the half a billion pounds worth of PPE has been used by the NHS, with much of it being unsuitable - but also at least some of those companies were picked because staffers were on the boards, and some of those companies were almost entirely fictional - just recently registered and with no reported accounts.
Here's a thread from one of the people at the Good Law Project explaining things.
Here's the thread detailing just how much money this fictional company made - it's upwards of £56 million for Andrew Mills and his wife. For unsuable PPE - the £56 million profit assumes that they had provided what the government asked for rather than substituting cheaper items in.
The Good Law Project has launched a judicial review claim (which they're crowd funding) to investigate the worst of the three so far.
This is the response from Pestfix
Posts
I'm getting serious "That Yellow Bastard" vibes from it
Like @klemming said, we largely switched to coursework
But it takes so long to mark - I miss exams!
Also, I am happy to report our students spend plenty of time in the library... Drinking and partying!
(really - turns out people were using the group study spaces for parties before the lockdown hit X)
I don't know how far you'd have to go back for that not to be true.
Students - far more traditionalist than you'd expect.
I was thinking more "evil vizier" and/or "that Wormtongue energy".
Reminds me a bit of The Hood from Thunderbirds or Ming the Merciless from Flash Gordon.
But yeah, definitely going for a 'power behind the throne' vibe.
I can sum it up too. Most deprived areas generally have greater number of BAME cohorts. Once it's broken down by ethnicity suddenly the SNP and the SQA are going to look 1000x worse than they already do.
The biggest fear is that when it's time for England's results I can see an even bigger fuckup coming.
I mean AQA are already sending letters to schools offering students unhappy with results or wanting better grades to take a fast track GCSE and A level exam throughout the autumn term. You know, year 11 and 13 students who have already left school/ college, to bring them back into already over capacity schools to take a 1 unit only exam. The same students who will be starting College or university in September..
If he doesn't do a steeped hands and say "Everything is proceeding as I have planned" in every appearance, that'll be a missed opportunity.
followed by someone pointing out a major flaw in his plan and him doubling down on it.
"My liege ... the satellites we purchased ... they're the wrong specs!"
"Everything. As. Planned.", he says, with a smile that's intended to intimidate but screams "I just shat myself".
Fortunately, this is the default look of Spitting Image puppets.
Yeah, evil vizier by way of Megamind
All that said it still makes more sense than "we give you a random grade based on how affluent your area is and if you don't like it tough titty".
I can't speak for any other universities, but right now the one I work at absolutely doesn't have the time or money to arrange entrance examinations. That's not a trivial task and we're already working at 150% in order to get (mostly?) online for Sept/October.
Should have you know, just trusted the teachers and the grades given by them as professionals.
It looks to me as if the teachers were just hopelessly optimistic regarding how well their most deprived students were going to do. It looks as if they overestimated the capabilities of all their students to pass exams by about 15%, and the algorithm corrected it based on the more accurate results from the previous years, giving everyone a generous ~2% bump in their chances of passing the exam (getting an A-C), this just looks worse for the most. deprived students. because of how overoptimistic their teachers were.
Now, you can clearly say that exams themselves have bias and aren't a good way to judge things, but if we are trying to say, "Noone did exams, what would have happened if they did" then I think the teachers making such awful predictions is the scandal, not the scores vs SIMD.
Except:
"If your school had no historical data, your teachers' estimates were just accepted. Astonishing."
The scandal is A. using estimates AT ALL, B. knocking down poor students so much worse. And you can see in the data posted that it wasn't anywhere near the same for everyone!
Worst off students:
85.1 to 69.9. 15.2 points, 82% of previous value
Best off students:
91.5 to 84.6. 6.9 points, 92% of previous value
The worst off students were hit much harder. This wasn't "oh everyone dropped by the same amount"
and of course grading students based on historical averages is just..what? You're telling me that a student that was doing perfectly before is going to fail now, for Reasons? Come on. You might as well draw grades out of a hat.
Why are we taking past exam results as reliable indicators of ability, when teachers and others have been screaming for years that they aren't?
It's simply not acceptable for students to get smarter, or teachers to get better at teaching a subject.
It's immoral and everyone involved should be fired out of a cannon into the sun.
Call me cynical but I've believed for years (based on the opinions of teachers who know about these things) that exams are a shit way of judging ability. Rich people love them because they're another system they've learned to subvert for their own benefit.
My parents (who aren't even rich) got me an English tutor before I took my highers. I ended up getting an A not because I was smart or talented or thoroughly understood the subject. I got an A because my folks hired a guy who essentially gave me the exam strategy guide on a plate. I went in there with a template in my head of what a grader will slap an A on. I sat the exams. I passed. It was forgotten forever a month later.
Exams test how good you are at exams. That's all.
https://tes.com/news/sqa-grilled-over-fairness-system-replacing-exams
This should highlight that MSP grilled the SQA overeats process back in May seeking reassurances that the SQA wouldn't just mark down due to the affluent nature of the schools area or its historical results. The SQA failed to address any of these.
Even so, I still can't believe that given a choice between everyone getting slightly better grades than they may have deserved or marking all the poor kids down to balance the stats this is what the SNP/SQA thought was the lesser evil. Sure if everyone got good grades you'd have a lot of right wing papers screeching about "muh meritocracy*", but you wouldn't have a full on uprising by parents and teachers in addition to the negative press. I don't know, does anyone think this system has any integrity at this point?
*Read: What's the point in rigging the game if there are no losers to make me look good?
Can't agree more!
I've gotten in trouble more than once as a teacher because my students "scored too highly" on an established exam.
What the...?
I'd work my ass off helping them prepare and they worked their asses off studying, taking practice tests, doing group revision sessions etc. I feel like, after students actually succeed and perform highly on a difficult exam, they should be praised for their hard work and not have their marks artificially lowered because "last year's cohort didn't do that well"
So, because of this kind of thing, I'm glad most of my classes are coursework-based now
(even though I know exams can be really useful at times and all that,
As mentioned earlier, given the two choices of marking pupils up, or marking pupils down, why is it seen to have been the better decision to mark them down? It seems like the negative consequences of that decision are far more impacting and so I'm just utterly confused by the decision here.
Well there are other factors that come in from trends with teacher awarding grades. Now I do not doubt that some schools will have teachers that vastly inflate marks for grades, but for the majority will be accurate. What teacher can't predict though are factors such as exam pressure, cramming, aspects of the test that an individual pupil might be stronger at than others, the testing environment in general, etc. All of which can have both negative and positive effects on grades. There will be countless students who are excellent throughout the years but struggle in a test environment and vice versa. Teachers can only go off what evidence they had.
What the SQA have done though is create a literal postcode lottery system with grades.
This is really confusing to me. Can you expand on this a little, like why specifically the management don't want good results?
To my understanding, it has to do with concerns about the increase in the number of first-class degrees awarded and grade inflation, like this: https://www.google.com/amp/s/amp.theguardian.com/education/2019/mar/24/universities--fines-too-many-top-degrees-grade-inflation
Does this rationale make sense to you?
Not to me...
I firmly believe my marking is fair and reliable and high marks are a result of our students' exceptional effort and not artificial grade inflation. Saying that students are only allowed to earn a small percentage of first-class degrees regardless of how well they perform seems crazy to me...
(but I started my teaching career in the US and it's possible I'm not judging the UK higher ed system fairly here - if so, please do let me know and I'll gladly change my views in response to new data ; )
You just need to add the fact that those in charge are evil, cruel, petty pieces of shit in a poorly constructed human suits into your calculations.
I can't speak about the details of your situation in particular, but it's basically a combination of trying to avoid degree inflation, and also attempting to maintain a level of intra- and inter-university reliability. A first from university X and university Y should signal roughly the same standard of work, as should a first in module A and module B in the same department.
If you're presenting your students the same (non-MCQ) exam as previous students received, but they're suddenly all receiving firsts, then sure, maybe it could be that they're just an exceptionally strong cohort or your teaching has particularly inspired them, but if they're not performing similarly well on other modules or exams then it's more likely to be the case that you're just marking them too leniently (especially since some of the pedagogical research I'm aware of suggests quite strongly that effort on individual modules will transfer - e.g., you shouldn't see a spike in attainment that only affects a single module if students are just studying harder). The general aim is to have some consistency.
As for whether temporary grade inflation at A level might impact universities... uh, maybe? It could be that some universities give out more conditional offers than they expect to see filled, and if so then they might find their resources stretched when they have more students in Sept/Oct than they're actually prepared for. But the sector's also expecting a drop in students actually wanting to go to university this year, so that might just wind up balancing things out. Possibly.
I think the potential effect on universities is mostly ignorable, and if there's any year where there should have been considerably more leniency than normal given to who gets to pass their exams, it should have been this year.
The issue is that the estimated vs historical data shows that the teachers were doing an awful job of predicting whether or not their students would pass exams (thats what this is. A predicted exam result)
So, the teachers saying " This student was doing well and would have gotten a grade A-C" is clearly weak information, whereas the historical data is well clustered and informative.
The fact that they just used the predictions if there was no historical data is clearly absurd, because the historical data shows the predictions are very poor.
I suppose that what this does show is that because the teachers were so bad at predicting, they should have just abandoned the whole exercise as they just didnt have the information they needed to make fair predictions for individuals, only for populations. And individuals are what matters here.
I also fear you are completely right about the real motivation behind such decisions @klemming
Thanks for the clear explanation @Burnage - I wish my EE explained it that way and we wouldn't have gotten into a huge fight about it X)
But I still agree 100% with @Santa Claustrophobia - curve grading is a loopy practice and should be stopped ; )
From what we're hearing from the Unis that are clients at work, most of them seem to be worrying about at the moment is (disregarding for the moment the practicalities of delivering teaching with covid, and international issues):
- oversubscription of conditional offers (most unis do offer more conditional places than they expect to be taken up), however this is always an issue. It's more of an issue if you aren't a prestige university and are less likely to be people's first choice, because you have to hedge against that by assuming a poorer rate of conversion from offer to enrolment.
- the fact that regardless of the grades awarded, this cohort is going to be of uncertain ability and there's an expectation of higher attrition as more people fail to progress. This is a big, big issue because the university loses those fees for the full duration of a (four year, in Scotland) degree
- some English students that have been offered places by Scottish unis are not going to be funded because the English finding bodies capped places at non-English unis after offers had been made
Interestingly, those unis that do contextual admissions will probably fare better, for a couple of reasons:
- contextual admissions assumes that pupils from deprived backgrounds are systematically undergraded, so they apply a calculation that is basically the inverse of what the sqa has done here, and make offers on the basis that a B from a "bad" school is worth more than a B from a "good" school
- those unis that do this well have made it work largely through investment in pastoral care and better mechanisms of support, which means they're better placed to keep students that struggle, and see lower attrition rates generally
There are different approaches and some of them are described on the UCAS website which had some case studies
I mostly know about it from having people that deal with admissions explaining it to me, but it's particularly a thing here (Scotland) because unis can secure additional "widening access" funding if they can show that they can deliver good academic outcomes for deprived groups as defined by the SIMD measures (basically you get more money the more you improve the academic achievement of the students you recruit)
The upshot is that there has been a fair amount of study of how best to do this, and one of the main takeaways is that issue of grading - a moderately achieving student from a poorly performing school is, on average, a more reliable bet for achieving a good degree than a moderately performing student from a high performing school
The tricky bit is that this is often only apparent if you correct for failures to complete a course of study that are non-academic in nature (medical, mental health, family circumstances, funding, etc), which is why you also need to do the pastoral bit and think carefully about how you allocate student support funding such as bursaries
Historical data doesn't signify what a current cohort would get. Results from schools fluctuate all the time. Have you considered factors like change in staffing? or Academy takeover? Was the cohort strong expectations this year or weaker than last year? Going on historical data is extremely foolish and has time and time again been discredited for having a poor reflection on a students actual results.
Have you considered the historical changing of grade boundaries to keep the top grades under a particular percentage by authorities like the SQA and JCQ?
Teachers were fine at predicting grades. They always have been.
Firstly, none of the half a billion pounds worth of PPE has been used by the NHS, with much of it being unsuitable - but also at least some of those companies were picked because staffers were on the boards, and some of those companies were almost entirely fictional - just recently registered and with no reported accounts.
Here's a thread from one of the people at the Good Law Project explaining things.
Here's the thread detailing just how much money this fictional company made - it's upwards of £56 million for Andrew Mills and his wife. For unsuable PPE - the £56 million profit assumes that they had provided what the government asked for rather than substituting cheaper items in.
The Good Law Project has launched a judicial review claim (which they're crowd funding) to investigate the worst of the three so far.
This is the response from Pestfix
And in unrelated news, after the last fiasco. It seems very likely that Dominic Cummings took a second trip up north.