# OCR Mathematics C ( Graduated Assessment)

Discussion in 'Mathematics' started by Lee Fields, Apr 20, 2011.

1. ### Lee FieldsNew commenter

A friend, whose daughter is doing her GCSE in this syllabus, has asked me to give her some extra support in preparation for the higher level terminal exam this June.
While I have used the textbooks, I have never taught the syllabus and I am finding difficulties understanding the assessment scheme. I wonder if a teacher familiar with the scheme can help.
From the specification document my understanding is
• 50% of the marks are allocated to the terminal exam which is in two equal parts - non calculator and calculator.
• The other 50% is gained from module exams. Ten modules are available, M1 (G grade standard) to M10 (A* standard). Each module is divided equally into a non-calculator and calculator part. Modules can be sat in January, March and June over Y10 & Y11. Each module can be repeated once prior to certification.
• Students (candidates!) would normally enter three appropriate modules the best two would be chosen for final grade.
• Here's where I get lost! How are the best two modules chosen if the candidate has attempted more than two modules over two years? For example if a candidate gains the (hypothetical) raw marks listed below, which two modules would be chosen and what overall grade might be awarded?
M7: 73%; M8: 71%; M9: 66%; Terminal Higher Level:
68%.

2. ### strawbsEstablished commenter

It's not done on raw marks - there is a scale of UMS equivalents whereby a certain % on a certain test gains a certain number of UMS points - then the "best2" are the two with the highest UMS points.
For the majority of candidates the highest 2 modules they take would provide the highest UMS as the points advance on a "ladder" - eg M7 is worth fewer points than M8 etc (there is a 50% overlap) - ie M7 overlaps 50% with the top half of M6 and 50% with the bottom half of M8.
So for a candidate who progresses well their M9 (unless they totally flunk it) will be worth more than their M7, regardless of whether the % is a bit less.

3. ### strawbsEstablished commenter

forgot to say kevchenko who posts on here is the OCR guru - hopefully he will pick up your post and correct anything I've got wrong above

4. ### kevchenko13

Hello Lee
In modular schemes, raw marks are converted to a uniform mark scale (UMS), which are pegged against the grades. So if a students gets 30 out of 50 on M10 it's worth more UMS than 30 out of 50 on M9, or M8, etc.
The Teacher's Guide will shed a bit more light on how it works. In your hypothetical situation, assuming the grade boudaries were all in roughly the same place on each module, then M8 and M9 would give the best two UMS scores and go forward with the terminal paper. Based on past results, those scores look quite close to the grade A borderline.
If you find out how many UMS she has for her modules already taken you can then see which two are the best. Adding these together and subtracting from the UMS required for her target grade (see Teacher's Guide p9) will tell you how many UMS she needs on the terminal. Then look that up on p8 to see what kind of performance that equates to.
There's a spreadsheet that does this sort of thing for you here but you'll need to know the UMS she already has, and approximate grade boundaries for the terminal paper (something like A* = 86, A = 70, B = 52, C = 35, D = 21, E = 13 won't be too far off based on past papers).
Hope this helps
Kevin Smith
OCR GCSE Mathematics Qualifications Manager

5. ### kevchenko13

Cross posted with strawbs!

6. ### strawbsEstablished commenter

and here is a copy of the Uniform marks table so you can see the "ladder" system I referred to
As a rough guide about 32% gets you the lowest "pass" for that particular module, about 60% is halfway up the range for that module, and about 92% would score you maximum UMS for that module.

7. ### strawbsEstablished commenter

good job I bigged you up then!

8. ### qwerty_tara

I can't believe I never saw that spreadsheet before! How fantastic... and how annoying we've only a month to go so I can't really use it!!!

9. ### kevchenko13

It's on the e-community. Register a school email address and you'll get email updates on when past papers etc are put up. Should still be useful I hope (are you not planning any January 2012 entries?).
Kevin

10. ### Lee FieldsNew commenter

Thanks Strawbs and kevchenko for the detailed and informative posts.
So, for example, a candidate scoring 32 / 50 raw marks on M8 would receive 146 Uniform Marks for that module ?
An issue that I'm perplexed about - a candidate scoring a very low mark on a High Module e.g. a score of 4 / 50 in M9 or M10. Surely this would not be awarded 143/144 Uniform Marks in M9 or 163/164 marks in M10?
Before the raw mark is mapped onto the Uniform Mark, is there prior adjustment of the raw mark depending on the national distribution of the marks for that paper in that sitting?

11. ### kevchenko13

It depends on where the grade thresholds are. If the grade B was 28/50 and grade C 15/50 then 32 marks would map to 146 UMS. But if the paper proved to be easier and the grade B threshold rose to 31/50, then 32 marks would score only 141 UMS.
No, it wouldn't. The lower grade threshold is mapped to the corresponding UMS and 0 marks always maps to 0 UMS. Taking M9 as an example, the B threshold is worth 140 UMS. If we say that it was set at 14 raw marks then each raw mark between 0 and 14 converts to 10 UMS, so 4 marks would score 40 UMS.
The two thresholds on each module (or three for M2 and M4) are set out of 50 by the awarding committee towards or at the end of the marking period. So yes, we are aware of the distribution of marks when the raw thresholds are set. As well as quantitative measures, they also use qualitative judgements in deciding where the thresholds should be set.