1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hi Guest, welcome to the TES Community!

    Connect with like-minded education professionals and have your say on the issues that matter to you.

    Don't forget to look at the how to guide.

    Dismiss Notice

Grade Boundaries for Maths GCSE 2017 Spec

Discussion in 'Mathematics' started by asymes86, Jul 4, 2016.

  1. DanRogan

    DanRogan New commenter

    We are just doing the Edexcel Autumn 2016 Mock set 1.
    Now that the papers are designed to have 50% of their content at grades 7-9, rather than the old spec which had 25% of marks at grades A-A*, might it not be reasonable to suggest that the marks needed for grade 7 would be a little under 50%? Why? Well on the old spec the mark needed for grade A was about 70% (compared with the 75% of marks available for questions below grade A.) So working with this figure and assuming candidates this year will be similar to those in previous years, we should not expect to have to get any grade 7-9 marks to achieve grade 7. That means I'd start looking for the grade 7 boundary around 47%.
    DamianR likes this.
  2. evivyover

    evivyover New commenter

    If you do the edexcel mock you can send your marks to thema and they will give you grade boundries. We are going to do this with our kids and just give them a percentage until Edexcel come back with the results. Edexcel need the 3 papers marks by 17th jan. It may be good for them so they can get feel of how students are resonding to the new papers...i would encourage people to do this...
  3. pentagonal

    pentagonal New commenter

    Edexcel will not be giving you any grade boundaries. They have clearly stated this.
  4. florence99

    florence99 New commenter

    We also sat all 3 secure Edexcel mocks and will be sending the results to Edexcel for the cohort analysis, however, as we have to set grade boundaries to give the students their results now I would be really interested to know where other schools have set. Would anyone be willing to share their ideas - I'll post our suggestions once all the papers have been marked.
    Also the deadline for mock mark submission is the 13th January to results plus, not the 17th, don't miss it!
    evivyover likes this.
  5. pentagonal

    pentagonal New commenter

    Grade 9 90%
    Grade 8 75%
    Grade 7 60%
    Grade 6 45%
    Grade 5 30%
    Grade 4 20%
    Grade 3 15%
  6. boznia

    boznia New commenter

    I am trying to make some of my own boundaries based on proportions of our school achieving certain grades in their mocks over the years, but I am not sure how to set the grade 4 anchor points. I wonder if you can help me.

    We do Edexcel, so higher and foundation tier.

    Normally 82% of our students get a C+ on their mocks. However some get it from Higher and some get it from Foundation and I don’t imagine the same proportions will be the case for the new more difficult papers ( especially as Higher is 50% A*/A questions ).
    So I am wondering how other people have set their grade 4 boundaries for each tier, based on the proportion of their students achieving a C.

    Hope someone can help.

  7. pi r squared

    pi r squared Occasional commenter

    In lieu of any other information, we can only assume the proportions of students achieving a Grade 4 in either paper will remain the same and base any estimated grade boundaries on this. Of course, they won't necessarily do this but we don't have crystal balls and can't predict the entry patterns up and down the country - but then, anything we come up with is just an educated guess, a shot in the dark, a **** into the wind, whatever metaphor you prefer.

    Even under the "old" system, there was no accepted logic as to which paper to enter the C/D borderline students into: some departments would swear blind that Higher was the best choice for "their" students, and others would believe the polar opposite. If there wasn't even a common train of thought in a system we've taught in for years, there are too many factors to possibly consider that it is worth applying Occam's razor - cut through the cruft and apply the simplest plausible system to create the fewest headaches.
  8. mathsman

    mathsman Occasional commenter

    A nice chap from Edexcel came to see us the other day and said the proportion of students who will get a grade 4 in the higher will be the same the proportion as who would normally get a grade C in the higher in the old specs.

    As he also said that half the questions in the higher paper will be A and A* questions

    I would say a grade 4 pass will need to be 20 to 25%
  9. pi r squared

    pi r squared Occasional commenter

    I'm not sure how, or why, they would make that claim because it just doesn't make any sense unless they're expecting there to be very little change in entry patterns. The "promise" has always been that broadly the same proportions of students who did get a C will now get a 4 overall. So if the numbers of students at each tier changes, the proportions of each tier getting a 4 cannot stay fixed or else the proportions getting a 4 overall will change.

    For example, my broad understanding from various bits of documentation AQA have produced is that approx. 60% of AQA candidates do Foundation, and approx. 40% of those get a C. 40% do Higher and approx. 85% get C or better. So that equates to around 58% of candidates getting a C+ overall. If under the new regime it became a 70/30 split, for example, but the tiered proportions remained the same, it would work out to be about 53% 4+ overall. It would also mean a large swathe of the Grade 5 kids would end up with a 4 instead (and so on up the chain) because of those 10% defecting to Foundation who would likely have been the "bottom" students on that paper.
  10. justmathssharing

    justmathssharing New commenter

  11. pi r squared

    pi r squared Occasional commenter

    The percentile ranks have now been added for the AQA Practice Paper Set 3 results. The Testbase Camp software is quite the pain in the backside to use and, as far as I can work out, the only way to get the percentile ranks is to select each student individually and get the rank for each student. Given that it literally takes 20 seconds to update each time you select a new student, and we're a year group of 240, this is certainly not the most efficient method to get what is pretty important information.

    Frustratingly there also doesn't seem to be any way of exporting the individual marks that staff have spent time and effort entering into a bespoke system. Usually we would use these to generate individual reports for students but it would have been a colossal waste of time to ask staff to enter the data into two different systems so we've been unable to do that this year.

    What I have done though is identified the percentile rank for the scores that we deemed to be our grade boundaries. Other schools will have used different ones but probably not a million miles away so the table below may still be useful. The "expected percentile rank" column is the percentile rank we would have expected given what we know about the distributions of results on the old spec in 2015 (I didn't have grade distributions per H or F for 2016). Obviously we do not know:
    • how representative the sample of schools that participated in the data drop are;
    • how grade distributions will change when the entry patterns change in terms of H and F;
    • how much progress students will make in raw marks between now and the Summer.
    Those caveats aside, here are my findings:

    On paper at least, this does suggest we have been quite conservative in our grade boundary setting and it is likely that we have been quite harsh in the results we have given students. I would certainly prefer it to be this way around than the other, but only seeing the actual grade boundaries in August is going to determine once and for all if we were.
  12. DeborahCarol

    DeborahCarol New commenter

    I was thinking the same. Haven't come across any bearings questions in Edexcel or AQA yet.
  13. montyzoomer

    montyzoomer New commenter

    I need some help with this if any one can. I'm the only maths teacher in a small SEN school and have 9 pupils. We are doing AQA foundation. Reading comments on here it seems that some think that 25% equates to a grade 3 or 4, whilst others think 31% is needed for a grade 1. I appreciate this is everyone's best guess, but with such a small sample of pupils I don't think I can make meaningful deductions. Please share your thoughts, thanks
  14. montyzoomer

    montyzoomer New commenter

    AQA set 1 paper 1 has a tiny bearing question on
  15. turnocka

    turnocka New commenter

    Based on our own data for 150 students doing an AQA mock, and using the 50% of the questions at grade C or above statement, I came up with these:
    50% - 4 40% - 3 30% - 2 20% - 1
    Similarly at higher tier:
    50% - 7 40% - 6 30% - 5 20% - 4
    This is more generous than most people seem to be giving, but it does fit in with the spread of data that pi r squared has posted. It also has the benefit of being easy for the students to latch onto.
    I definitely can't see 31% for a grade 1, as the new GCSE is significantly harder which means that the grade boundaries must be lower. (Unless of course Michael Gove is right, and asking harder questions will make the whole nation suddenly better at maths!)
  16. DeborahCarol

    DeborahCarol New commenter

    Thanks - yes, I noticed that recently. As you say, just a littlun. Haven't seen any trig-with-bearings...YET.

Share This Page