What’s the story with predicted grades anyway?
I’ve worked in the UK and Ireland so I know a little about predicted grades. They will always be contentious because they’re not a ‘real’ grade. However they have their uses which our current predicament with Covid-19 in Ireland has thrown up.
How do teachers come up with and use a predicted grade:
As a teacher, predicted grades require large amounts of ‘evidence’. Teachers (in the UK at least) get given vast amounts of educational data on a pupil at the beginning of any school year. This is often a full picture of the pupil’s attainment to date: from about the age of five up to and including GCSE results and CAT scores. This data is pre-generated and analysed by external agencies for you. It will tell you for example, what GCSE or A-Level grade the pupil is most likely to get in your subject. In Sixth Form (or in Ireland, Senior Cycle) this data is supposed to give you a very good idea, before you even start teaching them, of what the pupil will be able to do. I have some problems with this data because it puts pupils in a box before you ever teach them but that is a conversation for another day.
Teachers get asked about what they think Year 13 pupils are going to get relatively early in the school year (in October because pupils are doing their UCAS applications) so they start to build up a picture about a pupil very early in the year. They give a few ‘A-Level’ style assessments to get a clearer idea. The pre-existing data about the pupil is combined with these early assessments to help a teacher to ‘predict’ a grade. The same process is completed and revised throughout the year. Some courses will have a ‘coursework’ component (often a project worth about 20% or more) that the pupils work on independently with some teacher guidance.
Later in the school year, the pupils complete mock examinations (much like in Ireland) and this is further adds to the picture the teacher is completing about a pupil. Sometimes teachers are put under pressure from pupils, parents or even school leaders to change a predicted grade (because of a conditional university place offer – I’ll return to that later) but they don’t have to. I always made a prediction based on the available evidence I had, namely work that the pupils had done that I had marked using a marking scheme at the level they were taking the exam. By this I mean exam style work (done under exam conditions), coursework etc. Yes, the system is open to some tampering but I was trusted to be able to make those kind of judgement calls: I knew my pupils and crucially, I had the evidence to show what they could do.
What are predicted grades used for?
If a pupil was seriously ill or bereaved during the exam period, the predicted grades proved invaluable. This has happened to me a number of times and I had to submit my evidence about what grade I believed the pupil was most likely to get because they could not attend an exam for some valid reason. There is currently no similar system in Ireland though I know last year some steps have been taken to help pupils who are in difficult circumstances sit the state exams. However, currently most pupils have to resit the Leaving Cert exam the following year which is fair from ideal.
The other reason for the existence of predicted grades in the UK is the system of university entry there. The system is significantly different to the CAO process. UCAS requires students to apply to specific universities for specific courses early on. Universities have a good idea what grades students are likely to get so pupils don’t apply for something for which they are not suited or are unlikely to get the grades. Then the student could not get any offers at all. You can’t even apply to some of the more elite universities or high tariff courses (like medicine) if your GCSE results aren’t superb. Your attainment could have improved a lot since then (think of some lazy teenage boys) but you are still not considered. So what most pupils do is they hedge their bets: they have a ‘safe’ bet which requires lower grades than they hope they’ll get, something in the middle and then something aspirational; in case they do better than they expect.
Where it gets slightly more problematic is in the area of unconditional UCAS offers. I have known some pupils (especially in more recent years) to get offered places at UK universities based on predicted grades only; that means based on previous grades and predicted grades. So even if the pupil didn’t do as well as expected in the final exam, it wouldn’t matter. Their past performance was accepted as proof of their ability. This leads to obvious problems like pupils not really working for their last few months at school because their university place was in the bag. The reasons behind the recent enough spike in unconditional offers is just market forces I think; there were just too many university places for a few years so universities were keen to snap up ‘bright’ pupils early to get them to go there. This type of thing is not fair in my opinion but it does happen.
Problems with the use of predicted grades in Ireland:
A lot of accusations about predicted grades have been bandied around the media in Ireland in the past few weeks so I’m trying to address some of the issues. Yes, there are issues with predicted grades being used for the Leaving Certificate. The biggest one is that no one in Ireland really has any experience with them because they’ve never had to use them. Colleagues here don’t analyse exactly how their pupils do in the mocks for example, which is something I am now just in the habit of doing. As a teacher of MFL, I can tell pupils what their strongest skill is (speaking/listening) and where they need to improve to get a better result next time. Colleagues don’t necessarily keep very detailed records of exam style work, so they have no proof of what pupils can do.
There is also a complete lack of ‘coursework’ or equivalent in the Leaving Cert in its current form. Some practical subjects have some, but the bulk of subjects don’t so teachers are somewhat unfamiliar with applying end of course marking criteria to work on an ongoing basis. The new Junior Cert will go a long way to getting teachers up to speed with this. However my big bugbear with this is how very time consuming it is and in my experience, not necessarily of much benefit to students at all. Any decent teacher is giving student’s feedback on how to get better all the time anyway. Doing it in this ‘formal’ way adds a lot of bureaucracy, with not much tangible benefit in terms of outcomes. Students just don’t work harder and don’t do better if there is a coursework or continuous assessment element.
Parents and students are right to be sceptical about ‘hard’ markers and teachers with possible grudges or even unconscious bias because these things do exist. There is lots of statistical evidence to suggest varying kinds of unconscious bias on the part of teachers and this is very difficult to quantify when we have no previous data with which to compare it. In the UK, you are much more likely to get a lower prediction for example, if you from are black or from the Travelling community. Because we have never collected predicted grades before and been able to compare them with ‘real’ grades, we have absolutely no idea who these two numbers would compare which is crucial if the grade is to be worth anything and the integrity of the exam itself is to be maintained.
The best you can do is look at the evidence and try to use it to guide you. As a teacher I try to be ‘fair’ to every single student but that is hard. I know the kids I teach and I often (though not always) know who would do far better if they worked harder or that another is doing everything they can and eventually, they will get that H4 grade they really want.
What does it mean to be ‘fair’ anyway? But that’s a question for another day. 25/04/2020
Follow My Blog
Get new content delivered directly to your inbox.