[Misc] Exam results

[jackanada]

The point is that grades have been based on statistical modeling. I don't have full details but probably there is has been significant weighting to FFT (fisher family trust) modelling. This predicts attainment based on socio economic indicators. It is clearly more detailed but essentially rich professional parents nice neighborhood means you are far more likely to get top grades than poor parents in a slum. Obvious.
By using this data what has happened is that those poor kids who have worked hard and overcome their disadvantages to be on course for a top grade have been marked down, while those privileged students who have been managing to waste all their advantages and have done sweet fa for the best part of two years are not.
If you're bleating on about grade inflation, education gone soft, pointless subjects, snowflakes or whatever you are totally missing the point as to why this is such a travesty.

 

[Hastings gull]

They have just said on the news that in Wales, if the teachers grade assessments had been taken as the final grade, 40% of students would have got an A or A*. I am not saying that what has happened is correct, but having so many students getting top grades cannot be helpful, can it?

If you have been teaching nice, hard-working kids for 7 years and are then asked to estimate their grade, being aware of the significance for their uni aspirations, then the pressure on staff to err on the generous side is huge. I am not saying that this is necessarily ethical, but I know from 35 year's experience in secondary schools, that this is what will happen, given also the school's wish to look good with promising exam results. Governments know this and I assume that an attempt was made to lower the predictions somewhat, which I am sure will have been too generous. The TV reported that 30% of grades were affected.

 

[Hastings gull]

The point is that grades have been based on statistical modeling. I don't have full details but probably there is has been significant weighting to FFT (fisher family trust) modelling. This predicts attainment based on socio economic indicators. It is clearly more detailed but essentially rich professional parents nice neighborhood means you are far more likely to get top grades than poor parents in a slum. Obvious.
By using this data what has happened is that those poor kids who have worked hard and overcome their disadvantages to be on course for a top grade have been marked down, while those privileged students who have been managing to waste all their advantages and have done sweet fa for the best part of two years are not.
If you're bleating on about grade inflation, education gone soft, pointless subjects, snowflakes or whatever you are totally missing the point as to why this is such a travesty.

But it could well have been that staff in schools serving poorer areas might be precisely those desperate to see their students do well and are thus more likely to inflate their grades??

 

[Cheshire Cat]

When I did A levels back in the bad old days, the story was that the grade pass marks changed every year to ensure that similar percentages were awarded. So if 20% of pupils were intended to get A grades, the pass mark always matched the top 20%. One year a 75% pass would get you an A. The next year 80% was needed because the overall standard was better and the marking was higher. The top 20% were graded as A.

No idea if that was true or not.

 

[Pevenseagull]

The point is that grades have been based on statistical modeling. I don't have full details but probably there is has been significant weighting to FFT (fisher family trust) modelling. This predicts attainment based on socio economic indicators. It is clearly more detailed but essentially rich professional parents nice neighborhood means you are far more likely to get top grades than poor parents in a slum. Obvious.
By using this data what has happened is that those poor kids who have worked hard and overcome their disadvantages to be on course for a top grade have been marked down, while those privileged students who have been managing to waste all their advantages and have done sweet fa for the best part of two years are not.
If you're bleating on about grade inflation, education gone soft, pointless subjects, snowflakes or whatever you are totally missing the point as to why this is such a travesty.

may I use this comment verbatim elsewhere to get someone to shut the **** up?

I won't pass it off as my own work ... but that is so well written.

 

[jackanada]

But it could well have been that staff in schools serving poorer areas might be precisely those desperate to see their students do well and are thus more likely to inflate their grades??

As a general point scientific studies show that the wealthier you are the more likely you are to ignore rules. But here's a couple of examples of cheating at the end you seem to think would behave better.

None of the schools mentioned have had any of their students grades revised down this year.

https://www.theguardian.com/educati...-asks-teachers-to-exaggerate-exam-predictions

https://www.thesun.co.uk/uncategorized/4367710/private-school-cheating-scandal-inquiry/

 

[jackanada]

may I use this comment verbatim elsewhere to get someone to shut the **** up?

I won't pass it off as my own work ... but that is so well written.

You are most welcome to.
Probably won't help but such is the nature of trying to reason with the unreasonable.

 

[Pevenseagull]

But it could well have been that staff in schools serving poorer areas might be precisely those desperate to see their students do well and are thus more likely to inflate their grades??

Maybe the predicted grades will be revealed in due course.

In the interim there would appear to be a lot of kids who were like me 30 years ago who have been ****ed over by an algorithm.

 

[CliveWalkerWingWizard]

From what I can gather the algorithm is a lot more simple than some believe. On the surface it would appear to be fair if applied to all schools but when you think about it is is unfair to students from poorer backgrounds.

The system works like this:
The 3 year mean % is worked out for each grade, there is slight adjustment to reflect the GCSE qualifications of the most current cohort.

It favours independent/ higher achieving schools because due to the nature of their intake they will have more consistency in results.

So in an independent school let’s say each year every student gets an A, that will mean that this year every student gets an A.

Now let’s consider a school with a comprehensive intake, where some students are careers, some students need to do work 25 hours a week to support the family, some students have significant mental health issues and all the multitude of other issues that are more prominent in socioeconomic disadvantaged areas.

One year a cohort may have a significant number of students that will not achieve their potential by chance, let’s say every student gets an E, next year their is improvement and every student gets a C, the following year their is improvement again , possibly a combination of good teaching and just by chance less students with barriers to learning, every student gets an A. This year again is another year where every student is going to get an A so all the students are predicted A. So what happens? 33 % would get an A, 33% a C and 33% an E.

This is why the system is unfair. On top of that some students would have been hit 3 times so could be 3 grades down, is that fair? I do not know what the solution is, but it would have been fairer to take the best year in the last 3 as a benchmark. I am sure there were multiple brains sat in a room discussing how to award grades, but nobody could foresee the problem.

 

[Gwylan]

When I did A levels back in the bad old days, the story was that the grade pass marks changed every year to ensure that similar percentages were awarded. So if 20% of pupils were intended to get A grades, the pass mark always matched the top 20%. One year a 75% pass would get you an A. The next year 80% was needed because the overall standard was better and the marking was higher. The top 20% were graded as A.

No idea if that was true or not.

Yes, it was

See #118

20% of A levels weren't As though. I think something like 25%/30% had to fail too

 

[dsr-burnley]

From what I can gather the algorithm is a lot more simple than some believe. On the surface it would appear to be fair if applied to all schools but when you think about it is is unfair to students from poorer backgrounds.

The system works like this:
The 3 year mean % is worked out for each grade, there is slight adjustment to reflect the GCSE qualifications of the most current cohort.

It favours independent/ higher achieving schools because due to the nature of their intake they will have more consistency in results.

So in an independent school let’s say each year every student gets an A, that will mean that this year every student gets an A.

Now let’s consider a school with a comprehensive intake, where some students are careers, some students need to do work 25 hours a week to support the family, some students have significant mental health issues and all the multitude of other issues that are more prominent in socioeconomic disadvantaged areas.

One year a cohort may have a significant number of students that will not achieve their potential by chance, let’s say every student gets an E, next year their is improvement and every student gets a C, the following year their is improvement again , possibly a combination of good teaching and just by chance less students with barriers to learning, every student gets an A. This year again is another year where every student is going to get an A so all the students are predicted A. So what happens? 33 % would get an A, 33% a C and 33% an E.

This is why the system is unfair. On top of that some students would have been hit 3 times so could be 3 grades down, is that fair? I do not know what the solution is, but it would have been fairer to take the best year in the last 3 as a benchmark. I am sure there were multiple brains sat in a room discussing how to award grades, but nobody could foresee the problem.

No, it's not fair. Fairness is not possible. If a student's grade is wrong, it isn't fair, but there is no way to get a student's grade right.

On average, making each school's average mark similar to last year will get the greatest number of fair marks. That's the basis they marked it on. Yes, there will be schools where deadbeat teachers have left and been replaced by good ones; they will have unfairly low grades. And vice versa, schools where good teachers have left and been replaced by deadbeats will get unfairly high grades.

The problem with taking the best grades of the last three years is that there would inevitably be grade inflation. Anyone looking at the results of a 2020 child would know that the grades could not be relied on. Also unfair. This cannot be done fairly; all they can do is make their best effort.

 

[Bold Seagull]

No, it's not fair. Fairness is not possible. If a student's grade is wrong, it isn't fair, but there is no way to get a student's grade right.

On average, making each school's average mark similar to last year will get the greatest number of fair marks. That's the basis they marked it on. Yes, there will be schools where deadbeat teachers have left and been replaced by good ones; they will have unfairly low grades. And vice versa, schools where good teachers have left and been replaced by deadbeats will get unfairly high grades.

The problem with taking the best grades of the last three years is that there would inevitably be grade inflation. Anyone looking at the results of a 2020 child would know that the grades could not be relied on. Also unfair. This cannot be done fairly; all they can do is make their best effort.

Is it their ‘best’ effort though? I don’t think anyone is expecting 100% fairness, but As upgraded to A* at 4.7% in private schools but 0.3% in state? Come on, that isn’t someone making their best effort at fairness, that is an already skewed system skewing itself even more.

 

[timbha]

A stat flashed up yday that 28% of exams were awarded A or A* in England. Surely not?

 

[One Teddy Maybank]

No, it's not fair. Fairness is not possible. If a student's grade is wrong, it isn't fair, but there is no way to get a student's grade right.

On average, making each school's average mark similar to last year will get the greatest number of fair marks. That's the basis they marked it on. Yes, there will be schools where deadbeat teachers have left and been replaced by good ones; they will have unfairly low grades. And vice versa, schools where good teachers have left and been replaced by deadbeats will get unfairly high grades.

The problem with taking the best grades of the last three years is that there would inevitably be grade inflation. Anyone looking at the results of a 2020 child would know that the grades could not be relied on. Also unfair. This cannot be done fairly; all they can do is make their best effort.

Have to say this is how I see it.

Nothing is going to be ideal.......


Sent from my iPhone using Tapatalk

 

[Bold Seagull]

A stat flashed up yday that 28% of exams were awarded A or A* in England. Surely not?

49% of students privately educated got A - A* compared with 20% state schools. Private schools got the biggest boast of upgraded grades.

 

[timbha]

49% of students privately educated got A - A* compared with 20% state schools. Private schools got the biggest boast of upgraded grades.

Thanks. Does that mean the students got at least one grade A, or all As?

 

[Bold Seagull]

Thanks. Does that mean the students got at least one grade A, or all As?

I think it’s purely the number of A-levels taken that were awarded A or A*.

 

[timbha]

I think it’s purely the number of A-levels taken that were awarded A or A*.

Wow. There are some very smart kids around. I remember Grade C being regarded as good.

 

[beorhthelm]

No, it's not fair. Fairness is not possible. If a student's grade is wrong, it isn't fair, but there is no way to get a student's grade right.

On average, making each school's average mark similar to last year will get the greatest number of fair marks. That's the basis they marked it on. Yes, there will be schools where deadbeat teachers have left and been replaced by good ones; they will have unfairly low grades. And vice versa, schools where good teachers have left and been replaced by deadbeats will get unfairly high grades.

The problem with taking the best grades of the last three years is that there would inevitably be grade inflation. Anyone looking at the results of a 2020 child would know that the grades could not be relied on. Also unfair. This cannot be done fairly; all they can do is make their best effort.

no there probably isnt a completely fair way to sort this. the cries to revert to teacher's marks will lead to the most successfull cohort ever. the trouble is the moderation methods chosen by the exam boards are brutely simplistic and more unfair, judging a pupil their school. regrading will disproportinatly hit those in poorer performing schools who've worked hard and would have done well. the compromise seems sensible, take the mock result, as that should be free of teacher subjective view.

something we should be asking, is this how exam boards usually moderate exams with a simple algo based on school performance? have kids from poorer schools been consistently downgraded?

 

[Hampster Gull]

A stat flashed up yday that 28% of exams were awarded A or A* in England. Surely not?

I think the highest ever was around 28% so it’s not odd. I think the teach predictions were 38% for these two grades