It's strange now to think that my experience of applying is now out of date, and that the four interviews I had from 6 applications (Cambridge, UCL, Royal Holloway & Lancaster interviewed, Edinburgh and Durham did not) for Maths/Physics courses are no longer the norm.
I don’t know how long ago you started uni, but there are considerably more freshers these days than even a decade ago. My university’s intake target (I.e. the maximum the government would allow us) was over 6,000.
In some cases it would just mean using a bigger lecture theatre.
Assuming such are available. We’re in the middle of a massive building project because we don’t have enough theatres big enough for our normal population, never mind for a suddenly-doubled one.
Edit to add: and of course that building project has been the first thing cancelled in response to the predicted loss of income next year...
My university had about 3 places that would fit everyone doing first year maths, computer science or engineering. If you're skipping rows as well as seats, rooms that would fit 100 are down to 20.
Most, if not all, university courses for this year will be conducted using technology as far as possible. Lecture theatre space will be irrelevant as they won't be used, students will attend lectures broadcast across the universities intranet. Most tutorial and seminar groups will use Zoom. Practical sessions, such as labs in sciences, will need to be in person - but, you can have less people in each lab if you have more flexibility in timetables not dominated by lectures at specific times.
I wonder how many universities return to traditional lectures. With online lectures you can expand the number of students without needing to build more and bigger lecture theatres.
So if you just allow the results to stand, then more students will go to the better univwersities this year than in normal years, but those universities should have the flexibility to allow this.
That would be fine, if it weren’t for the fact that the government, in its wisdom, decided to impose a student number control (i.e. a maximum number of students we can recruit) on us this year. In theory this was to protect the worse universities from potential failure due to all the better universities over-recruiting UK students to make up for their predicted drop in overseas ones.
If the government now said: 'Oops, the cap was a silly idea' and removed it, and then just allowed predicted grades to stand, how much of a problem would that be?
Now that confirmation and clearing has finished, massive. You’d suddenly have a whole bunch of new applicants on the market, which is fine, but for those unis who want to maintain their average entry grades* few of them would be good enough. But there would be good enough students in lesser units who would suddenly be in a position of being able to ditch the uni they’ve already agreed to go to in favour of a better one.
Being brutal, but if they couldn't hack it and dropped out (or dropped back to the lesser uni), wouldn't the problem then solve itself?
It would be chaos. You’d effectively be starting the whole confirmation and clearing process again, but with no time to plan or prepare.
How much time do you normally have? Did you know before Thursday which applicants had met their offer?
Given the circumstances being based from home has a lot of merit (although paradoxically it might lead to more travel).
The trouble is, the systems set up so most people value degree results more from Oxford&Cambridge more than Northampton (historically out of legally enforced jealousy).
As well as larger lecture halls etc, where would the students live?
If students went to the nearest university, most of them could live at home.
In Scotland it seems to be much more common to attend a university in the same city than in England where attending a more distant university is the norm. That might be a result of the vast majority of people in Scotland living in or very near one of two cities. I think this does mean that the social aspects of university, the first time living away from home with all the responsibilities and opportunities that brings, can be missed. Though, most of the students we have out doing project work with us have moved into digs with mates by the time they've got to 3rd or 4th year. But, even if they leave home, students who go to the local university will still miss the experience of a different culture and new places, and will potentially miss out on new friends if they maintain their network of friends from school, continue at the same church etc.
Besides, the local university may well not be the best place to study their subject of choice. Or, there may not be a local university (when I went, the nearest universities to me were in London, which would have taken me more than an hour to get to from home).
It's strange now to think that my experience of applying is now out of date, and that the four interviews I had from 6 applications (Cambridge, UCL, Royal Holloway & Lancaster interviewed, Edinburgh and Durham did not) for Maths/Physics courses are no longer the norm.
I don’t know how long ago you started uni, but there are considerably more freshers these days than even a decade ago. My university’s intake target (I.e. the maximum the government would allow us) was over 6,000.
It would be chaos. You’d effectively be starting the whole confirmation and clearing process again, but with no time to plan or prepare.
How much time do you normally have? Did you know before Thursday which applicants had met their offer?
To the second question, yes. We get the A Level results in advance, under strict embargo. We know how many students we have with firm acceptances before we go anywhere near clearing, which is how we know how many places we have to fill on any given course.
To the first, we start planning for the confirmation and clearing process months in advance. Even the logistics of getting enough people on the phone lines takes a ridiculous amount of time.
It would be chaos. You’d effectively be starting the whole confirmation and clearing process again, but with no time to plan or prepare.
How much time do you normally have? Did you know before Thursday which applicants had met their offer?
To the second question, yes. We get the A Level results in advance, under strict embargo. We know how many students we have with firm acceptances before we go anywhere near clearing, which is how we know how many places we have to fill on any given course.
I have read that teachers have been over predicting for a number of years. They should calculate the average percentage of over predicting for the past 5 years and use that percentage for reducing the predictions this year.
I have read that teachers have been over predicting for a number of years. They should calculate the average percentage of over predicting for the past 5 years and use that percentage for reducing the predictions this year.
Except that the level of over-prediction will vary substantially from teacher to teacher, and there is a difference between offering a predicted grade for UCAS (who specifically ask you to provide a reasonable best-case-scenario prediction) and attempting to project what a student will actually get. Plus it's the case that over-prediction is not universal - if you've got 10 candidates whose most likely result is a B you'll likely find that 6 get a B, 2 get a C, one gets a D and one gets an A. If you look at predicted grades from modelling they're always expressed as percentages. It's entirely normal to see a candidate who has (say) a 40% chance of a C and a 35% chance of a D. If you've got three candidates like that, which one do you give a D to? What you're suggesting is not that far from what's actually taken place. The problem is that in attempting to replicate nationally the normal pattern of results they've treated a lot of individuals unfairly. There's no good solution, but the least-worst one is that followed in Scotland and France - yes you get a year of higher grades, but with all the other issues school leavers are going to face in the current economic climate? Barely a blip.
With online lectures you can expand the number of students without needing to build more and bigger lecture theatres.
But can you still justify big fees?
Can you ever justify charging students fees? But, that's a different discussion.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home? With a large class like that there's no real space for personal interaction, no real option for a student to raise their hand or come forward at the end and say "I don't understand". With recorded lectures, there's always the option to rewind and listen again to see if it's clearer the second time around (which could make that a better option).
I know the course my head of group delivers already includes lecture material being placed online, with the result that several of the students who've signed up for the course are never seen as they use the online material (and, as a masters course this includes some of the overseas students who have paid fees). So, there's anecdotal evidence that at least some students would consider quality on-line lectures to be better value than the current pattern of only having the Powerpoint file online. At present, students who don't attend lectures and then complain that the Powerpoint isn't good enough don't get a particularly favourable response if they complain. Obviously if all the material is online then that does need to be good material, as it is the course material rather than a supplement to attending lectures.
Open University fees aren’t much cheaper than bricks and mortar units these days, despite being mostly online (they look cheaper because they are based on part time study). The prices were increased in 2012 when our funding changed and we were allowed to use student loans. Obviously our courses are purposely written for online delivery but other universities also have online courses already and I assume they are similarly priced.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home? With a large class like that there's no real space for personal interaction, no real option for a student to raise their hand or come forward at the end and say "I don't understand". With recorded lectures, there's always the option to rewind and listen again to see if it's clearer the second time around (which could make that a better option).
than a supplement to attending lectures.
Only 100? 55 or more years ago, English I and Psychology I each had enrolments of around 700 at the uni I attended - tutorials were a much more sensible size fortunately.
With online lectures you can expand the number of students without needing to build more and bigger lecture theatres.
But can you still justify big fees?
Can you ever justify charging students fees? But, that's a different discussion.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home?
The ability of the students to mix with each other afterwards, sorting themselves into groups based on various levels of perceived keenness, the shared struggles in the libraries afterwards and then a few brave groups approach the lecturer with feedback which feeds into subsequent lectures.
A little bit of light relief in the midst of the turmoil is provided by this comment (from a reader) on the Guardian website:
I tried using the supercomputer in the maths department to run my own algorithm on the government, but no matter what parameters I try to use every single member of the cabinet gets a fail grade.
It's the use of averages that's the problem, isn't it?
If only. The algorithm is coming out with average results for state schools that are lower than one would expect based on past performance by the schools. My attempts to work out what exactly is being done are coming across qualified statisticians saying, I don't quite follow this bit.
There are also further meannesses, where the spread of results is skewed downward.
It's the use of averages that's the problem, isn't it?
If only. The algorithm is coming out with average results for state schools that are lower than one would expect based on past performance by the schools. My attempts to work out what exactly is being done are coming across qualified statisticians saying, I don't quite follow this bit.
There are also further meannesses, where the spread of results is skewed downward.
There some explanation and exploration here, with an accompanying interactive chart.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home? With a large class like that there's no real space for personal interaction, no real option for a student to raise their hand or come forward at the end and say "I don't understand". With recorded lectures, there's always the option to rewind and listen again to see if it's clearer the second time around (which could make that a better option).
than a supplement to attending lectures.
Only 100? 55 or more years ago, English I and Psychology I each had enrolments of around 700 at the uni I attended - tutorials were a much more sensible size fortunately.
I said "100+", as an arbitrary number. Smaller classes provide more opportunity for eye contact with the lecturer, and for the lecturer to judge how well the class seem to be taking in the material and adjust the rate of presentation and/or go over things a second time if they see a lot of confused faces. I don't think there's a definite number that get too large for that - for a start it'll depend on the lecturer and how good they are at reading body language, and the nature of the lecture theatre (is it too large and everyone's at the back? how's the lighting? etc). That level of feedback, obviously, disappears in pre-recorded lectures.
The social side of lectures, as @chrisstiles noted, is important too. But, that could be replicated to an extent by students gathering on Zoom to talk through the lecture they've all just watched on their computers.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home? With a large class like that there's no real space for personal interaction, no real option for a student to raise their hand or come forward at the end and say "I don't understand". With recorded lectures, there's always the option to rewind and listen again to see if it's clearer the second time around (which could make that a better option).
than a supplement to attending lectures.
Only 100? 55 or more years ago, English I and Psychology I each had enrolments of around 700 at the uni I attended - tutorials were a much more sensible size fortunately.
I said "100+", as an arbitrary number. Smaller classes provide more opportunity for eye contact with the lecturer, and for the lecturer to judge how well the class seem to be taking in the material and adjust the rate of presentation and/or go over things a second time if they see a lot of confused faces.
I'm not sure our lecturers noticed when people fell asleep.
The social side of lectures, as @chrisstiles noted, is important too. But, that could be replicated to an extent by students gathering on Zoom to talk through the lecture they've all just watched on their computers.
In general I'm a big booster of technology, but having sat through more than my fair share of video conferences I don't think it really can - conferencing doesn't really allow for the spontaneous breakout sessions that come out of a group of people meeting in real life, and have to be moderated far more to be successful (which in turn needs someone with a grasp of the subject).
I've done online courses that contained a discussion aspect using forum/video conferencing technology and even at their best they were a very poor substitute for meeting in person.
The phrase there was "to an extent", in other words I agree with you. But, unless there's a return to full lockdown, there's still nothing stopping groups of students meeting over coffee (or, possibly more likely, beer) to talk about the lectures when they're pre-recorded.
I'm waiting to hear back (at a regular Zoom chat) from our current PhD student who has just "attended" a conference at MIT that was re-arranged for video conference to see how that worked out for her. I can see how the talks could work fairly well (even if some of them ran into the wee small hours for those in the UK), but the important aspects of a conference are the informal chats over coffee, lunch, dinner, in the bar ... it'll be interesting how those worked.
I said "100+", as an arbitrary number. Smaller classes provide more opportunity for eye contact with the lecturer, and for the lecturer to judge how well the class seem to be taking in the material and adjust the rate of presentation and/or go over things a second time if they see a lot of confused faces. I don't think there's a definite number that get too large for that - for a start it'll depend on the lecturer and how good they are at reading body language, and the nature of the lecture theatre (is it too large and everyone's at the back? how's the lighting? etc). That level of feedback, obviously, disappears in pre-recorded lectures.
The social side of lectures, as @chrisstiles noted, is important too. But, that could be replicated to an extent by students gathering on Zoom to talk through the lecture they've all just watched on their computers.
I know it was arbitrary and agree with your other comments. The real teaching was done in the much smaller tutorials.
There's a potentially good discussion about how valuable a lecture is in teaching. I'd agree that the real teaching happens elsewhere - in tutorials or labs, in students working together going over lecture material, working through problems, delving into text books and other resources. And, if they're not that valuable, does the pandemic forcing universities into alternatives to the in-person lecture present an opportunity to re-define education away from lectures to other methods that actually work? If someone else doesn't get around to it first, I might start such a discussion in Purgatory when I've a bit more time this evening.
And, what we say about lectures may also apply to sermons ...
@Telford it is certainly true that teachers over-predict. But as I described in a previous post, this is partly because UCAS say that teachers are meant to be somewhat optimistic when assigning "predicted grades", that predicted grades are supposed to be "aspirational" without being "unrealistic". Obviously this is rather subjective and unsatisfactory and leads to grade inflation. As a teacher you feel that if a pupil outperforms their predicted grades you've done them a serious disservice, whereas if they underperform that's only to be expected in a good proportion of cases.
The teacher-calculated grades this year were meant to be accurate rather than aspirational. The calculated grades we assigned this April were certainly much lower overall than the UCAS predicted grades we submitted last November. I was quite fierce like this and reluctantly knocked down a number of grades during internal moderation on the grounds that it was unrealistic to expect our cohort to be much better than average this year. I know that colleagues at other schools said they had not done this and that it was up to the government to moderate them down if they had a problem with it.
So I will be really quite irritated if the government does a U-Turn. I will feel that my honesty has been penalised and the dishonesty of others has been rewarded.
As a teacher you feel that if a pupil outperforms their predicted grades you've done them a serious disservice, whereas if they underperform that's only to be expected in a good proportion of cases.
That sounds like special pleading to me - esp as the figures seem t suggest that 75% of grades were over inflated last year.
It's an awful thing and a massive cock up but it has also revealed for the first time, some very disturbing evidence about internal gradings. That isn't doing anyone any favours.
As a teacher you feel that if a pupil outperforms their predicted grades you've done them a serious disservice, whereas if they underperform that's only to be expected in a good proportion of cases.
That sounds like special pleading to me - esp as the figures seem t suggest that 75% of grades were over inflated last year.
A percentage likely not helped by it being so early in the run of the new-style A-Levels. I know marking them (maths) it was very different from the old system. I wouldn't be surprised if that was a contributing factor. The problem with UCAS predicted grades is that everyone knows that they're what get students offers, and they know that you're more likely to get a place if you miss your offer by a grade or two than if your predicted grades are a grade or two less than the potential offer in the first place. Add in that UCAS predicted grades are done more than 6 months out from the exams and you've got a recipe for great uncertainty and strong encouragement to err on the side of generosity.
When teachers are asked for "best case realistic" predictions they'll give that - what they think students will achieve if they work diligently over the next 6 months (for the UCAS submission) and if they don't fall ill on exam day or are otherwise unlucky. Probably most students in the stress of the exam don't quite achieve their potential and these predictions will be higher than the results, some students may work their socks off and get lucky with questions on topics they've revised hard and exceed the predictions. There are lots of stats floated about 75% of predictions in previous years being inflated, but it's never stated by how much ... a prediction of marks a few percent above the achieved mark could still be a grade different.
If teachers are asked in April to give their predictions, even on the same "best case realistic" criteria, those would be different from the predictions previously submitted to UCAS. There's less uncertainty about how hard the students will work between the prediction and the exam (though, of course, this year there was the massive uncertainty from working at home), there may have been mock exams taken which would provide additional information for a prediction, etc. In a normal year, predictions made in April (though, I'd be surprised if they were normally made then) will be more accurate than those made in September. The two sets of predictions are simply different and it would be inappropriate to use them as a simple comparison to scale one set against the other, even if teachers were working on the same instructions on how to make predictions (eg: both are to be accurate or both aspirational). You might as well ask a farmer to predict his apple harvest from data about orange harvests.
I wonder how many universities return to traditional lectures. With online lectures you can expand the number of students without needing to build more and bigger lecture theatres.
Will be interesting. Despite my distaste for 9am lectures (I went to a few, but mostly when I hadn't been to bed yet), I far prefer in-person lectures to zoom ones - zoom just feels dead by comparison to a full lecture theatre.
For those who like interactive lectures, with all the students having clickers to select their choice of answer, you can do that with the various electronic classroom solutions, but I wonder how efficient they are.
(Whether I went to a lecture mostly depended on the lecturer - a few were fantastic, some were OK, and some were dire. In the latter category I'll include the poor bloke who was so terrified of giving a lecture that he kept permanent contact with his security overhead projector. His feet would wander around at the front of the class, but his hand remained firmly clamped on the OHP.)
At present, students who don't attend lectures and then complain that the Powerpoint isn't good enough don't get a particularly favourable response if they complain.
This is where I point out that a set of powerpoint slides to be used as an accompaniment to a lecture and a set of powerpoint slides intended to be readable by themselves are completely different beasts. And for an undergraduate lecture, the latter usually already exists, in convenient bound form, accompanied by a number of questions to help the student ensure that they have understood the material.
We had some lecturers who would deliberately hand out incomplete printouts of the slides, in the hope that this would force students to actively take notes, and so pay attention. I have no idea whether it worked or not.
When teachers are asked for "best case realistic" predictions they'll give that - what they think students will achieve if they work diligently over the next 6 months (for the UCAS submission) and if they don't fall ill on exam day or are otherwise unlucky. Probably most students in the stress of the exam don't quite achieve their potential and these predictions will be higher than the results, some students may work their socks off and get lucky with questions on topics they've revised hard and exceed the predictions. There are lots of stats floated about 75% of predictions in previous years being inflated, but it's never stated by how much ... a prediction of marks a few percent above the achieved mark could still be a grade different.
If teachers are asked in April to give their predictions, even on the same "best case realistic" criteria, those would be different from the predictions previously submitted to UCAS. There's less uncertainty about how hard the students will work between the prediction and the exam (though, of course, this year there was the massive uncertainty from working at home), there may have been mock exams taken which would provide additional information for a prediction, etc. In a normal year, predictions made in April (though, I'd be surprised if they were normally made then) will be more accurate than those made in September. The two sets of predictions are simply different and it would be inappropriate to use them as a simple comparison to scale one set against the other, even if teachers were working on the same instructions on how to make predictions (eg: both are to be accurate or both aspirational). You might as well ask a farmer to predict his apple harvest from data about orange harvests.
I think it's increasingly clear that there is no appropriate data-set to use to calibrate teacher's predictions. Thus the assertion that 75% of predictions are wrong or whatever is the latest government talking point is entirely meaningless.
So we have an algorithm that is making a correction for something that may or may not be there (and more importantly may be significantly uneven) plus other adjustments which clearly disadvantage some groups of students more than others.
It's insane. This system is clearly not defensible. And having left it this late, no doubt it is having a massive effect on university places and messing up Clearing no end.
And the bottom line in all this is that real people's lives are being messed with by our constantly feckless government.
I think it's increasingly clear that there is no appropriate data-set to use to calibrate teacher's predictions. Thus the assertion that 75% of predictions are wrong or whatever is the latest government talking point is entirely meaningless.
Yeah, I think this is true. We can be pretty certain that, over the nation as a whole, this cohort is not appreciably different from any other (so when the average teacher grades come in a grade higher than the real results from last year, we know that's the average error). But that says nothing at all about the distribution of individual errors.
You can have the average prediction be off by a grade with 100% of the predictions being wrong, or with as little as about 30% of the predictions being wrong (technically less than that even, but let's assume some vague rationality in the predictions), and there is no good data to tell those scenarios apart.
The actual algorithm (thanks to whoever posted the link upthread) is completely nuts. If I had been involved with the group developing the algorithm, and that was the conclusion the group had come to, I'd be removing my name from the paper.
It's becoming more clear to me that you simply cannot generate a reasonable set of grades for this cohort that are comparable to last year's grades, or next year's grades. So give up on it. That means your only problem is what to do with this year's university admissions. (Yes, there's another issue over GCSEs - whether someone has a pass in English and Maths is a magic status indicator, but there's generally more time to fix that.)
I don't know much about current university admissions in the UK. I assume that universities tend to over-offer, by some factor that depends on their data from previous years, to account for some people failing to make the grade, and others being accepted by a higher-ranking institution. So I do wonder how over-subscribed each course at each institution would be if we just assumed that every applicant had achieved their offer.
As a teacher you feel that if a pupil outperforms their predicted grades you've done them a serious disservice, whereas if they underperform that's only to be expected in a good proportion of cases.
That sounds like special pleading to me - esp as the figures seem t suggest that 75% of grades were over inflated last year.
Given the (lack of) consistency in marking revealed in that Ofqual paper, I'm unclear why this would be the particular inaccuracy you'd fixate above.
And as I keep saying in this thread; the figures - where adjusted -- rely on the teachers being perfectly accurate in ranking students.
And WHERE is our Lord Protector in all this? Still trying to put up a tent in Scotland?
I hope it pours with rain, blows a gale, and makes the bloody thing collapse around him (without affecting those others unfortunate enough to be in the contaminated area, of course).
Once again, The Celtic Fringe gets it right first...
Success! U turn means teacher assessments will be used in England.
Let’s hope the next announcement is a removal of the cap on university places so there is room for everyone who has now been awarded the grades they needed.
Comments
My university had about 3 places that would fit everyone doing first year maths, computer science or engineering. If you're skipping rows as well as seats, rooms that would fit 100 are down to 20.
I wonder how many universities return to traditional lectures. With online lectures you can expand the number of students without needing to build more and bigger lecture theatres.
Being brutal, but if they couldn't hack it and dropped out (or dropped back to the lesser uni), wouldn't the problem then solve itself?
How much time do you normally have? Did you know before Thursday which applicants had met their offer?
If students went to the nearest university, most of them could live at home.
The trouble is, the systems set up so most people value degree results more from Oxford&Cambridge more than Northampton (historically out of legally enforced jealousy).
Besides, the local university may well not be the best place to study their subject of choice. Or, there may not be a local university (when I went, the nearest universities to me were in London, which would have taken me more than an hour to get to from home).
Still more recent than me!
To the second question, yes. We get the A Level results in advance, under strict embargo. We know how many students we have with firm acceptances before we go anywhere near clearing, which is how we know how many places we have to fill on any given course.
To the first, we start planning for the confirmation and clearing process months in advance. Even the logistics of getting enough people on the phone lines takes a ridiculous amount of time.
Ah thanks, that's the bit I was missing.
Except that the level of over-prediction will vary substantially from teacher to teacher, and there is a difference between offering a predicted grade for UCAS (who specifically ask you to provide a reasonable best-case-scenario prediction) and attempting to project what a student will actually get. Plus it's the case that over-prediction is not universal - if you've got 10 candidates whose most likely result is a B you'll likely find that 6 get a B, 2 get a C, one gets a D and one gets an A. If you look at predicted grades from modelling they're always expressed as percentages. It's entirely normal to see a candidate who has (say) a 40% chance of a C and a 35% chance of a D. If you've got three candidates like that, which one do you give a D to? What you're suggesting is not that far from what's actually taken place. The problem is that in attempting to replicate nationally the normal pattern of results they've treated a lot of individuals unfairly. There's no good solution, but the least-worst one is that followed in Scotland and France - yes you get a year of higher grades, but with all the other issues school leavers are going to face in the current economic climate? Barely a blip.
At the end of the day, what's functionally different from a lecturer standing in front of a lecture theatre of 100+ students delivering a lecture and delivering the same material to a camera that the same 100+ students watch at home? With a large class like that there's no real space for personal interaction, no real option for a student to raise their hand or come forward at the end and say "I don't understand". With recorded lectures, there's always the option to rewind and listen again to see if it's clearer the second time around (which could make that a better option).
I know the course my head of group delivers already includes lecture material being placed online, with the result that several of the students who've signed up for the course are never seen as they use the online material (and, as a masters course this includes some of the overseas students who have paid fees). So, there's anecdotal evidence that at least some students would consider quality on-line lectures to be better value than the current pattern of only having the Powerpoint file online. At present, students who don't attend lectures and then complain that the Powerpoint isn't good enough don't get a particularly favourable response if they complain. Obviously if all the material is online then that does need to be good material, as it is the course material rather than a supplement to attending lectures.
Partially,
but it's more the use of non averages some schools are notably up, the college's as mentioned are down 20,,%
Only 100? 55 or more years ago, English I and Psychology I each had enrolments of around 700 at the uni I attended - tutorials were a much more sensible size fortunately.
The ability of the students to mix with each other afterwards, sorting themselves into groups based on various levels of perceived keenness, the shared struggles in the libraries afterwards and then a few brave groups approach the lecturer with feedback which feeds into subsequent lectures.
I tried using the supercomputer in the maths department to run my own algorithm on the government, but no matter what parameters I try to use every single member of the cabinet gets a fail grade.
There are also further meannesses, where the spread of results is skewed downward.
"Teachers are inaccurate at grading students" sits badly alongside "Teachers can perfectly rank their students".
There some explanation and exploration here, with an accompanying interactive chart.
The social side of lectures, as @chrisstiles noted, is important too. But, that could be replicated to an extent by students gathering on Zoom to talk through the lecture they've all just watched on their computers.
I'm not sure our lecturers noticed when people fell asleep.
In general I'm a big booster of technology, but having sat through more than my fair share of video conferences I don't think it really can - conferencing doesn't really allow for the spontaneous breakout sessions that come out of a group of people meeting in real life, and have to be moderated far more to be successful (which in turn needs someone with a grasp of the subject).
I've done online courses that contained a discussion aspect using forum/video conferencing technology and even at their best they were a very poor substitute for meeting in person.
I'm waiting to hear back (at a regular Zoom chat) from our current PhD student who has just "attended" a conference at MIT that was re-arranged for video conference to see how that worked out for her. I can see how the talks could work fairly well (even if some of them ran into the wee small hours for those in the UK), but the important aspects of a conference are the informal chats over coffee, lunch, dinner, in the bar ... it'll be interesting how those worked.
I know it was arbitrary and agree with your other comments. The real teaching was done in the much smaller tutorials.
And, what we say about lectures may also apply to sermons ...
The teacher-calculated grades this year were meant to be accurate rather than aspirational. The calculated grades we assigned this April were certainly much lower overall than the UCAS predicted grades we submitted last November. I was quite fierce like this and reluctantly knocked down a number of grades during internal moderation on the grounds that it was unrealistic to expect our cohort to be much better than average this year. I know that colleagues at other schools said they had not done this and that it was up to the government to moderate them down if they had a problem with it.
So I will be really quite irritated if the government does a U-Turn. I will feel that my honesty has been penalised and the dishonesty of others has been rewarded.
It's an awful thing and a massive cock up but it has also revealed for the first time, some very disturbing evidence about internal gradings. That isn't doing anyone any favours.
A percentage likely not helped by it being so early in the run of the new-style A-Levels. I know marking them (maths) it was very different from the old system. I wouldn't be surprised if that was a contributing factor. The problem with UCAS predicted grades is that everyone knows that they're what get students offers, and they know that you're more likely to get a place if you miss your offer by a grade or two than if your predicted grades are a grade or two less than the potential offer in the first place. Add in that UCAS predicted grades are done more than 6 months out from the exams and you've got a recipe for great uncertainty and strong encouragement to err on the side of generosity.
If teachers are asked in April to give their predictions, even on the same "best case realistic" criteria, those would be different from the predictions previously submitted to UCAS. There's less uncertainty about how hard the students will work between the prediction and the exam (though, of course, this year there was the massive uncertainty from working at home), there may have been mock exams taken which would provide additional information for a prediction, etc. In a normal year, predictions made in April (though, I'd be surprised if they were normally made then) will be more accurate than those made in September. The two sets of predictions are simply different and it would be inappropriate to use them as a simple comparison to scale one set against the other, even if teachers were working on the same instructions on how to make predictions (eg: both are to be accurate or both aspirational). You might as well ask a farmer to predict his apple harvest from data about orange harvests.
Will be interesting. Despite my distaste for 9am lectures (I went to a few, but mostly when I hadn't been to bed yet), I far prefer in-person lectures to zoom ones - zoom just feels dead by comparison to a full lecture theatre.
For those who like interactive lectures, with all the students having clickers to select their choice of answer, you can do that with the various electronic classroom solutions, but I wonder how efficient they are.
(Whether I went to a lecture mostly depended on the lecturer - a few were fantastic, some were OK, and some were dire. In the latter category I'll include the poor bloke who was so terrified of giving a lecture that he kept permanent contact with his security overhead projector. His feet would wander around at the front of the class, but his hand remained firmly clamped on the OHP.)
This is where I point out that a set of powerpoint slides to be used as an accompaniment to a lecture and a set of powerpoint slides intended to be readable by themselves are completely different beasts. And for an undergraduate lecture, the latter usually already exists, in convenient bound form, accompanied by a number of questions to help the student ensure that they have understood the material.
We had some lecturers who would deliberately hand out incomplete printouts of the slides, in the hope that this would force students to actively take notes, and so pay attention. I have no idea whether it worked or not.
I think it's increasingly clear that there is no appropriate data-set to use to calibrate teacher's predictions. Thus the assertion that 75% of predictions are wrong or whatever is the latest government talking point is entirely meaningless.
So we have an algorithm that is making a correction for something that may or may not be there (and more importantly may be significantly uneven) plus other adjustments which clearly disadvantage some groups of students more than others.
It's insane. This system is clearly not defensible. And having left it this late, no doubt it is having a massive effect on university places and messing up Clearing no end.
And the bottom line in all this is that real people's lives are being messed with by our constantly feckless government.
AFZ
True, alas, and there seems to be b*gger-all we can do about it...
https://bbc.co.uk/news/live/world-53802215
The gurgling noise you can hear is the *government* going down the plug-hole after yet another U-turn...
Yeah, I think this is true. We can be pretty certain that, over the nation as a whole, this cohort is not appreciably different from any other (so when the average teacher grades come in a grade higher than the real results from last year, we know that's the average error). But that says nothing at all about the distribution of individual errors.
You can have the average prediction be off by a grade with 100% of the predictions being wrong, or with as little as about 30% of the predictions being wrong (technically less than that even, but let's assume some vague rationality in the predictions), and there is no good data to tell those scenarios apart.
The actual algorithm (thanks to whoever posted the link upthread) is completely nuts. If I had been involved with the group developing the algorithm, and that was the conclusion the group had come to, I'd be removing my name from the paper.
It's becoming more clear to me that you simply cannot generate a reasonable set of grades for this cohort that are comparable to last year's grades, or next year's grades. So give up on it. That means your only problem is what to do with this year's university admissions. (Yes, there's another issue over GCSEs - whether someone has a pass in English and Maths is a magic status indicator, but there's generally more time to fix that.)
I don't know much about current university admissions in the UK. I assume that universities tend to over-offer, by some factor that depends on their data from previous years, to account for some people failing to make the grade, and others being accepted by a higher-ranking institution. So I do wonder how over-subscribed each course at each institution would be if we just assumed that every applicant had achieved their offer.
https://bbc.co.uk/news/uk-53810655
Where's the hollow laughter emoji when you need it?
Given the (lack of) consistency in marking revealed in that Ofqual paper, I'm unclear why this would be the particular inaccuracy you'd fixate above.
And as I keep saying in this thread; the figures - where adjusted -- rely on the teachers being perfectly accurate in ranking students.
I hope it pours with rain, blows a gale, and makes the bloody thing collapse around him (without affecting those others unfortunate enough to be in the contaminated area, of course).
Once again, The Celtic Fringe gets it right first...
Let’s hope the next announcement is a removal of the cap on university places so there is room for everyone who has now been awarded the grades they needed.