Research shows teachers over-generous for 2020 grades.

The FFT Education Datalab (who specialise in educational research) have published some incredibly interesting research regarding the submitted centre assessment grades that are going to be the basis of all given A-level and GCSE grades this summer. See the full report here.

After receiving the preliminary GCSE centre grades from 1,900 schools, (over half of all schools) they then compared these results to that of the final results of 2019 and the findings are very interesting:

•

This year's teacher-assesed grades are higher than those awarded in 2019 exams. In every subject looked at, the average grade proposed from 2020 was higher than the average grade awarded in 2019.

•

In most subjects this difference was between +0.3 and +0.6

•

Looking across all subjects, if these grades were given out this summer then you would see the share of grade 9s increase from 4.8% of all grades awarded to 6.3%. The share of results receiving a grade 7 or above would increase from 23.4% of all grades to 28.2%, while the share of results receiving a grade 4 or above would increase from 72.8% to 80.7%.

What could this mean for students grades this summer

Although there are some caveats to this data (detailed below) this means it is very likely that the GCSE grades submitted to the exam boards are significantly above those that were awarded last year.

This leaves Ofqual and the Exam Boards with a problem, and most likely means that will have to apply significant statistical moderation to the grades submitted by schools, bringing them down on average. It is therefore likely that the rank orderings that schools were required to put students in will have an impact and be used to shift some pupils down from one grade to the next.

Why do Ofqual and Exam Boards need to adjust grades at all?

It is not my place to say if this is correct or not, but Ofqual and Exam Boards have decided they have to statistically moderate grades given this summer for two reasons:

1.) To help correct any significant variation of teacher/centre being too generous or too harsh on their marking. So by applying a form of statistical moderation to the grades they can help limit this to a certain degree.

2.) To validate grades given this year compared to years gone by or years to come. If grades are to vary too significantly this year, it would bring students achievements this academic year into question and would invalidate these grades in the eyes of some employers/universities etc.

There are probably two reasons that teacher assessed grades are higher than the grades students got last year (Thanks to @SarcAndSpark)

1) Exams using 9-1 grades have only been around for 2 years in most subjects. This doesn't give teachers, even experienced ones, that much data to go on. It's likely grade boundaries would have gone up again in most subjects this year, so it's hard to know what a grade 7/8/9 student really looks like. Teachers have got better at teaching the course with experience, so it's likely this year's students are better at new style exams than those in previous years.

2) Psychologically, it's quite difficult for a teacher to fail someone if they think there's a chance they might have passed. This will probably impact students on the 3/4 and 4/5 borderline the most. If a student could have got a 5, for example, then it's very hard for a teacher to give them a 5, knowing that might mess them up for next year. Schools didn't get lots of guidance from exam boards, so I don't think it's a huge surprise that grades aren't exactly the same as previous years!

Things worth keeping in mind about the above data

•

The grades received by FFT Education DataLabs here are the preliminary grades the schools had organised. It is possible that grades were adjusted by centres after given to DataLabs and before being finalised.

•

This is for GCSE grades only, although you can make up your own mind if a similar pattern will be seen at A-level (I imagine it will personally) it is worth noting that GCSE grades have their own unique characteristics.

•

Data only makes up about half of the overall picture, so although the general trend is likely to be correct, to what extent is still to be seen.

I thought it was best to share this research with everyone, if you have any questions or comments please feel free to post them below, I will try to answer them as best as I can.

Reply 1

tacobeth

not gonna lie, this makes me even more anxious now for results day

Reply 2

Evil Homer

TSR Community Staff

Original post by tacobeth

not gonna lie, this makes me even more anxious now for results day

Hey

I understand how this thread could have that impact, but we think it's important to share all the information we can about what's going on this year. I hope you understand.

Can I ask specifically why this has you more anxious?

Reply 3

tacobeth

Original post by Evil Homer

Hey

just worried they’ll place more focus now on looking at previous attainment when calculating our grades, can’t really do much now but it’s good you posted the research!

Reply 4

TCL

Original post by tacobeth

just worried they’ll place more focus now on looking at previous attainment when calculating our grades, can’t really do much now but it’s good you posted the research!

They have said that they are only looking at previous attainment at cohort level, not for individuals, so if you have made a lot of progress since Y7 that will not affect the way YOUR grades are moderated.

Reply 5

TCL

Original post by Evil Homer

........
2) Psychologically, it's quite difficult for a teacher to fail someone if they think there's a chance they might have passed. This will probably impact students on the 3/4 and 4/5 borderline the most. If a student could have got a 5, for example, then it's very hard for a teacher to give them a 5, knowing that might mess them up for next year. Schools didn't get lots of guidance from exam boards, so I don't think it's a huge surprise that grades aren't exactly the same as previous years!
.........

This will be something which happens at all grade boundaries. If a student consistently scores on the borderline, roughly 50% 6s, 50% 7s, or 50% 7, 50% 8, the teacher will want to assume they would have had a good day for the exam. In a normal year I think most students find that some of their grades will be as expected, some better, some worse, particularly if they generally hover around grade boundaries and have a tendency to make silly mistakes/misread questions.