Predicted Grades Appeals

JPF82 · 15-08-2020 9:57am #1

With all the goings on in the UK system with the predicted grades, the below piece from the journal.ie came to my attention. https://www.thejournal.ie/leaving-cert-results-9-5175809-Aug2020/

In particular, this bit

The Department of Education confirmed to TheJournal.ie that students would be able to view their Calculated Grade broken down into the teacher-assessed grade and ranking.

“The results of the Calculated Grades will issue to students via the Calculated Grades Student Portal and simultaneously to schools on Monday, 7 September 2020,” it said.

“On Monday 14 September the student portal will open for appeal applications and at that time all students will be able to see a record of their individual estimated percentage mark and ranking assigned to them by their school.

I thought the students would only be able to see a ranking, and not the grade I assigned them. It's not that I wouldn't stand over my grades.

While looking at this, I believe RTÉ have a news article saying students only see the ranking. It could just be the journal being wrong.

km79 · 15-08-2020 10:33am

Just on the LC
September 7th is creeping closer ......what’s happening !
Online only ? Come in to schools while current students are there ?

RealJohn · 15-08-2020 10:47am

JPF82 wrote: »

I thought the students would only be able to see a ranking, and not the grade I assigned them. It's not that I wouldn't stand over my grades.

I’d rather they saw the mark I assigned them than their rank in the class. The person I assigned last place to is definitely likely to say I had something personal against them.

Treppen · 15-08-2020 11:24am

RealJohn wrote: »

I’d rather they saw the mark I assigned them than their rank in the class. The person I assigned last place to is definitely likely to say I had something personal against them.

Good point, I can see zero benefit to anyone with revealing a ranking and teachers predicted grade, just give them the final grade and let the department take the flack. This is just a recipe for disaster and really bad feelings towards teachers, especially in a small close knit community.

Dunno about ye but I'm enjoyed wearing a face mask at the local shopping centre passing by students who don't recognise me.

am_zarathustra · 15-08-2020 11:47am

Does anyone know who actually did the analysis? Since they announced I've been keeping an eye out but haven't seen anything. I can't see anyone in the Department being capable of this kind of statistical analysis or writing an algorithm so I assume an outside agency was contracted? The SEC might have been able to just about do it but they rightly washed their hands of the process

mirrorwall14 · 15-08-2020 12:26pm

am_zarathustra wrote: »

Does anyone know who actually did the analysis? Since they announced I've been keeping an eye out but haven't seen anything. I can't see anyone in the Department being capable of this kind of statistical analysis or writing an algorithm so I assume an outside agency was contracted? The SEC might have been able to just about do it but they rightly washed their hands of the process

I believe that at least four have been transferred in from the sec

am_zarathustra · 15-08-2020 1:11pm

Four people? I can't imagine that being enough. I assume it's an automated algorithm designed to pick anomalies? The SEC would hardly have a couple of programmers on staff, though I could be wrong!

Fincool · 15-08-2020 1:45pm

Students will see the ranking and the percentage that was assigned to them by their teacher.

mtoutlemonde · 15-08-2020 2:14pm

am_zarathustra wrote: »

Four people? I can't imagine that being enough. I assume it's an automated algorithm designed to pick anomalies? The SEC would hardly have a couple of programmers on staff, though I could be wrong!

With the introduction of online marking, the SEC have recruited a number of people in the past few years that are available for all IT related problems. They have a solid knowledge in the systems so I assume their knowledge of the marking system and bell curves was required. I remember seeing a post for IT director (title may not be correct) advertised a few years ago with a salary far bigger than a teachers.

am_zarathustra · 15-08-2020 4:16pm

That's true but I suppose they bought in the online correcting system wholesale so they needed support as opposed to development which would be required for this. I'm mostly curious, the weighting of the algorithm will decide people grades, I'd be interested to see which factors carried what weight.

Mardy Bum · 15-08-2020 4:33pm

This will go one of two ways:
A) Grades are decided largely by algorithm with the teacher ranking being the most important ingredient.

All grades match the teachers' grades.

Both will be a disaster as proved by the English and Scottish systems. Teachers in both countries handed out marks to the point that there were twice as many achieving the highest grade in contrast with last year. This led to the algorithm deciding the grade which had a drastic effect on high achieving students in schools with low results over the past few years.

It is a lose lose situation.

shesty · 15-08-2020 4:48pm

The "bell curve" is applied in performance management in many workplaces - 3 of my workplaces to date have used it, where employees recieve a "mark" for the year's work against certain goals, the marks all then go in for a company-wide review carried out by Senior Management, where every single mark is reviewed and "standarised" across the company, and then you are told how you did, and dependent on that bonuses/payrises (company dependent).

I cannot describe the fighting it causes.The complaining and moaning and appealing and whining -and these are adults, doing mostly the same jobs year in, year out across a wide variety and level of roles. I can only imagine how the LC grades situation is going to go.

Millionaire only not · 15-08-2020 5:23pm

My understanding is that u will only see the ranking and teacher grade only if your appealing.
The appeals process is only that everything was done correctly marks will not change from what the teacher gave .
That’s my understanding which may not mean much .

am_zarathustra · 15-08-2020 5:25pm

Mardy Bum wrote: »

This will go one of two ways:
A) Grades are decided largely by algorithm with the teacher ranking being the most important ingredient.
All grades match the teachers' grades.

Both will be a disaster as proved by the English and Scottish systems. Teachers in both countries handed out marks to the point that there were twice as many achieving the highest grade in contrast with last year. This led to the algorithm deciding the grade which had a drastic effect on high achieving students in schools with low results over the past few years.

It is a lose lose situation.

I heard anecdotally that some schools had grades sent back for review, I don't know how true that is.

It's hard to know which way they will go. Another option is don't touch DEIS grades and bell curve the rest, that might stop the headlines similar to Scotland but it's open to its own issues.

There isn't a good answer, my guess is the LC will happen this year "by hook or by crook"

Treppen · 15-08-2020 6:40pm

Mardy Bum wrote: »

This will go one of two ways:
A) Grades are decided largely by algorithm with the teacher ranking being the most important ingredient.
All grades match the teachers' grades.

Both will be a disaster as proved by the English and Scottish systems. Teachers in both countries handed out marks to the point that there were twice as many achieving the highest grade in contrast with last year. This led to the algorithm deciding the grade which had a drastic effect on high achieving students in schools with low results over the past few years.

It is a lose lose situation.

Minister Norma Foley said Socio Economic status and Gender would be accounted for!

Surprised she left out Learning Difficulty... Although not really as they are usually consigned to the forgotten afterthought in education.

"the outcomes need to be reviewed using different demographic characteristics which will include gender and socio-economic status to ensure that the grades are as fair and equitable as possible."

https://www.gov.ie/en/press-release/7777b-minister-foley-announces-next-steps-for-this-years-leaving-certificate-students/

Moody_mona · 15-08-2020 6:51pm

JPF82 wrote: »

With all the goings on in the UK system with the predicted grades, the below piece from the journal.ie came to my attention. https://www.thejournal.ie/leaving-cert-results-9-5175809-Aug2020/

In particular, this bit

The Department of Education confirmed to TheJournal.ie that students would be able to view their Calculated Grade broken down into the teacher-assessed grade and ranking.

“The results of the Calculated Grades will issue to students via the Calculated Grades Student Portal and simultaneously to schools on Monday, 7 September 2020,” it said.

“On Monday 14 September the student portal will open for appeal applications and at that time all students will be able to see a record of their individual estimated percentage mark and ranking assigned to them by their school.

I thought the students would only be able to see a ranking, and not the grade I assigned them. It's not that I wouldn't stand over my grades.

While looking at this, I believe RTÉ have a news article saying students only see the ranking. It could just be the journal being wrong.

This was not my understanding at all, that students would see grade or rank?

Treppen · 16-08-2020 4:12pm

Moody_mona wrote: »

This was not my understanding at all, that students would see grade or rank?

I emailed asti Union for clarification and got nothing, although they might have been out of office.

soso02 · 16-08-2020 5:54pm

Surely under GDPR etc, students should have access to the school grade & ranking and the department's percentage mark ?

I know of a student who was only contacted by the CGEO this week to be told it was not possible to receive a calculated grade. Only sorting this out now, no wonder for the delay.

Does the student who took the minister to the high court have a case ? Hopefully this will be nipped in the bud to deter students from doing likewise..

Do the department expect schools to provide support to students who wish to sit the exams ? Their term was cut short by three/four weeks, how can they be expected to sit exams having been out of school for 7/8 months. I'm not confident anyways the exams will be in November given the SEC's overly cautious approach thus far.

Given results are available from 9am on the portal, I don't think many would go into school anyways.

Would it not make more sense to follow England's approach and have appeals be done on a school level. It won't be just one student being downgraded ? How would people feel about being able to choose between the calculated grade or mock result ? Happened in England, and our calculate grades system bears a striking resemblance to England's.

Treppen · 17-08-2020 11:20am

soso02 wrote: »

Surely under GDPR etc, students should have access to the school grade & ranking and the department's percentage mark ?

I know of a student who was only contacted by the CGEO this week to be told it was not possible to receive a calculated grade. Only sorting this out now, no wonder for the delay.

Does the student who took the minister to the high court have a case ? Hopefully this will be nipped in the bud to deter students from doing likewise..

Do the department expect schools to provide support to students who wish to sit the exams ? Their term was cut short by three/four weeks, how can they be expected to sit exams having been out of school for 7/8 months. I'm not confident anyways the exams will be in November given the SEC's overly cautious approach thus far.

Given results are available from 9am on the portal, I don't think many would go into school anyways.

Would it not make more sense to follow England's approach and have appeals be done on a school level. It won't be just one student being downgraded ? How would people feel about being able to choose between the calculated grade or mock result ? Happened in England, and our calculate grades system bears a striking resemblance to England's.

It all goes back to the fact that papers for mocks were available , so anyone student cogging would be rewarded over say a student who's teacher set their own particularly hard mock paper for a very good class.

Even with a choice approach the cheating student would be rewarded with a course place possibly taking it from an honest student.

In terms of precedent, we know there are currently 5th and 2nd year students getting wise to predicted grades and questioning grades from teachers this summer. Going forward it would be in everyone's best interests to do as well as possible in the mocks for 2021... By simple hard work or other nefarious means.

It would make a mockery of the mocks.

Mardy Bum · 17-08-2020 12:19pm

Treppen wrote: »

It all goes back to the fact that papers for mocks were available , so anyone student cogging would be rewarded over say a student who's teacher set their own particularly hard mock paper for a very good class.

Even with a choice approach the cheating student would be rewarded with a course place possibly taking it from an honest student.

In terms of precedent, we know there are currently 5th and 2nd year students getting wise to predicted grades and questioning grades from teachers this summer. Going forward it would be in everyone's best interests to do as well as possible in the mocks for 2021... By simple hard work or other nefarious means.

It would make a mockery of the mocks.

It has become clear that calculated grades don't work because lots of teachers find it impossible to be objective and are too supportive and also the algorithm doesn't work. Entrance exams for courses would be a much better option. It would also encourage students to consider their choices wisely.

mirrorwall14 · 17-08-2020 12:28pm

GCSE calculated grades in the North pulled. So far the A levels still stand....

doc_17 · 17-08-2020 12:40pm

They are probably going to revert to the teacher predicted grades in England now as well (announcement at 4pm).

There’s going to be some panic in the Dept now! And it’s putting teachers in a hard place as well. Some will have done it given the benefit of the doubt to all their student, some won’t have. It’s going to be a sh*t storm.

Mardy Bum · 17-08-2020 1:36pm

doc_17 wrote: »

They are probably going to revert to the teacher predicted grades in England now as well (announcement at 4pm).

There’s going to be some panic in the Dept now! And it’s putting teachers in a hard place as well. Some will have done it given the benefit of the doubt to all their student, some won’t have. It’s going to be a sh*t storm.

As usual the teachers who do their job properly get crucified and those who don't will get away with it..

RealJohn · 17-08-2020 2:25pm

soso02 wrote: »

How would people feel about being able to choose between the calculated grade or mock result ? Happened in England, and our calculate grades system bears a striking resemblance to England's.

It would be terrible, but I could see it happening, and I don't think it would be any worse than what's supposed to be happening anyway (which is also terrible, in case that wasn't clear).

Mardy Bum wrote: »

It has become clear that calculated grades don't work because lots of teachers find it impossible to be objective and are too supportive and also the algorithm doesn't work.

Agreed on teachers. Teachers, in general, are inconsistent, at best, in all aspects of examining. That's why the SEC has marking conferences and advising examiners - even with a marking scheme, teachers don't mark consistently, and that's without having any bias involved.
The algorithm doesn't work as a concept, but that's more down to the unpredictability of any one student's exam performance.

Mardy Bum wrote: »

Entrance exams for courses would be a much better option. It would also encourage students to consider their choices wisely.

Well yes, for those going to college. It doesn't solve the problem of students not having a leaving cert though (which is essentially what's happening now, since the leaving cert they're going to be given will be worthless).

rainbowtrout · 17-08-2020 9:18pm

Moody_mona wrote: »

This was not my understanding at all, that students would see grade or rank?

Well they would have to see grade and rank, otherwise they have nothing to appeal on.

E.g. student receives a H4 in English from DES in three weeks time and the mark was 62. Is that the same as the teacher predicted grade or not? Hears that other students in the class did better than them despite the fact that they know they were top in English all year. Appeal can only establish if process was adhered to correctly, and presumably part of that includes data entry.

Student looks at teacher mark and sees that it was a 92 and they were ranked 1 in the class. Appeal reveals that data was incorrectly entered and someone hit a 6 instead of a 9. Student is successful in appeal.

Case 2: Student gets H4 (again 62) from DES. Checks teacher grade and discovers that it was also a H4, only a 65. Mark was moderated but overall grade did not change. Student knows that an appeal will not change the mark. Whether they feel hard done by or not is another thing.

Moody_mona · 17-08-2020 9:27pm

rainbowtrout wrote: »

Well they would have to see grade and rank, otherwise they have nothing to appeal on.

E.g. student receives a H4 in English from DES in three weeks time and the mark was 62. Is that the same as the teacher predicted grade or not? Hears that other students in the class did better than them despite the fact that they know they were top in English all year. Appeal can only establish if process was adhered to correctly, and presumably part of that includes data entry.

Student looks at teacher mark and sees that it was a 92 and they were ranked 1 in the class. Appeal reveals that data was incorrectly entered and someone hit a 6 instead of a 9. Student is successful in appeal.

Case 2: Student gets H4 (again 62) from DES. Checks teacher grade and discovers that it was also a H4, only a 65. Mark was moderated but overall grade did not change. Student knows that an appeal will not change the mark. Whether they feel hard done by or not is another thing.

Thanks Rainbowtrout. I suppose I thought it would be more DES based. As per your case A, students receives 62%. Isn't happy and appeals. Department check that the grade awarded was correct, it wasn't, they update it. Similar to Junior Cert appeals of years gone by I guess?

Treppen · 17-08-2020 9:30pm

rainbowtrout wrote: »

Well they would have to see grade and rank, otherwise they have nothing to appeal on.

E.g. student receives a H4 in English from DES in three weeks time and the mark was 62. Is that the same as the teacher predicted grade or not? Hears that other students in the class did better than them despite the fact that they know they were top in English all year. Appeal can only establish if process was adhered to correctly, and presumably part of that includes data entry.

Student looks at teacher mark and sees that it was a 92 and they were ranked 1 in the class. Appeal reveals that data was incorrectly entered and someone hit a 6 instead of a 9. Student is successful in appeal.

Case 2: Student gets H4 (again 62) from DES. Checks teacher grade and discovers that it was also a H4, only a 65. Mark was moderated but overall grade did not change. Student knows that an appeal will not change the mark. Whether they feel hard done by or not is another thing.

Say the Student Gets H1 from the teacher at 91% ( the student's own Junior cert result was exceptional 100%) and finds the department has dropped them to H2.

They appeal and find there's no clerical error, the algorithm has performed it's grade adjustment accordingly, what then ?

rainbowtrout · 17-08-2020 10:16pm

Moody_mona wrote: »

Thanks Rainbowtrout. I suppose I thought it would be more DES based. As per your case A, students receives 62%. Isn't happy and appeals. Department check that the grade awarded was correct, it wasn't, they update it. Similar to Junior Cert appeals of years gone by I guess?

I presume so, based on the fact that they've said the appeals were essentially an administrative check. Otherwise anyone that was downgraded from their teachers grade would appeal.

Treppen wrote: »

Say the Student Gets H1 from the teacher at 91% ( the student's own Junior cert result was exceptional 100%) and finds the department has dropped them to H2.

They appeal and find there's no clerical error, the algorithm has performed it's grade adjustment accordingly, what then ?

Talk to Joe I guess. Sit the exams in November if they want. The majority of people will still be offered a college place and then the fuss will die down.

There will be a furore in the media in a few weeks time I presume, but if the grades are in line with previous years then the algorithm has done it's job.

That's something I'd be curious to know about the English/Scottish grades. Aside from the meltdown about grades being downgraded over there, I'd like to know how the moderated grades compared to the bell curve of previous years and what the bell curve was like for the teachers' predicted grades.

But to come back to your example, context is important. Are they the only H1 predicted for that class/school. If so and the school usually gets a H1 in that subject, then it appears to be unfair. But if the school typically gets three H1s per year in the subject and 5 students were awarded H1s by the teachers of that subject and this student was either in 4th or 5th position in the ranking of those H1s, then the students who scored higher H1s will get the grade, which is what would happen in the exams. The curve would look after it.

I had three excellent students in my LC class last year, any one of them could have got a H1 on the day. I typically would only have one H1 when I do have one. I would have been rightly stuck this year if I was trying to predict grades for those three to separate them out. I viewed their scripts with them last summer. All three got H2s, but they were A2s in old money. An 85, 87 and 88 so really there was nothing between them, and that happens across the country every year, only I would suspect teachers would have given the benefit of the doubt in the predictions for many of these and given them the H1. For the 87/88 one extra part of a question answered correctly would have carried them over the line. Even one extra H1 per school doesn't look like much to the teacher awarding it but the cumulative effect is quite considerable.

Mardy Bum · 17-08-2020 10:26pm

rainbowtrout wrote: »

I presume so, based on the fact that they've said the appeals were essentially an administrative check. Otherwise anyone that was downgraded from their teachers grade would appeal.

Talk to Joe I guess. Sit the exams in November if they want. The majority of people will still be offered a college place and then the fuss will die down.

There will be a furore in the media in a few weeks time I presume, but if the grades are in line with previous years then the algorithm has done it's job.

That's something I'd be curious to know about the English/Scottish grades. Aside from the meltdown about grades being downgraded over there, I'd like to know how the moderated grades compared to the bell curve of previous years and what the bell curve was like for the teachers' predicted grades.

Classes with over 15 were moderated solely by the algorithm which used school profile data i.e. the teacher grade wasn't used at all. Classes with between 5-15 had used a mixture of the teacher grade and school profile data. Less than five just used the teacher grade.

rainbowtrout · 17-08-2020 10:35pm

Mardy Bum wrote: »

Classes with over 15 were moderated solely by the algorithm which used school profile data i.e. the teacher grade wasn't used at all. Classes with between 5-15 had used a mixture of the teacher grade and school profile data. Less than five just used the teacher grade.

Thanks for that. That's interesting. The algorithm wasn't applied consistently then. That process to my mind would appear to skew in favour of fee paying schools where class sizes are smaller, outcomes are generally good, students have all the advantages. Students in those schools (the Etons and Harrows of the world) do achieve good grades, but they have everything going in their favour and they were always likely to have good grades predicted for them too and in this system, not have them moderated. Students in a large comprehensive are always going to be fighting to get their deserved grade when they are being graded on their socio economic status only.

Mardy Bum · 17-08-2020 11:50pm

rainbowtrout wrote: »

Thanks for that. That's interesting. The algorithm wasn't applied consistently then. That process to my mind would appear to skew in favour of fee paying schools where class sizes are smaller, outcomes are generally good, students have all the advantages. Students in those schools (the Etons and Harrows of the world) do achieve good grades, but they have everything going in their favour and they were always likely to have good grades predicted for them too and in this system, not have them moderated. Students in a large comprehensive are always going to be fighting to get their deserved grade when they are being graded on their socio economic status only.

That is exactly what happened. The first issue was wild over predictions which then led to teacher predictions being ignored in larger cohorts which was obviously stupid at the same time. Even the best schools had a large minority of grades decreased. Some comprehensive schools had 60+% downgraded because of the algorithm though. Now they are reverting to the best of the teacher grade, mock grade or calculated grade. An omnishambles which N Foley may have ahead of her.

Predicted Grades Appeals

Comments