New Accountability – thoughts and questions

UPDATE – based on the Update on Progress 8 measure and Stuart Locks’s blog (

Since the @headsroundtable meeting with @timleunig  I had been asking questions of what the new progress measure means to my school.  A few things I hadn’t been able to get right in my head and I couldn’t work out why.

Over time a couple of comments and assumptions started to get me thinking about the curriculum we are about to offer students in year:

  1. The measure appears to be based on APS with the total based on the basket divided by 8 (see my assumptions below.  So if only 7 subjects count it is still divided by 8. – because of the double count the actual final will be divided by 10 (if a student completes Maths & English and 6 others that count)
  2. The calculations threw an oddity with the double count of English & maths. It appears that a student that gets 8 Cs but will count as a B (and so on in all grades) for accountability, as long as there are no Ebacc gaps – see above
  3. It does appear that @headteachers’ blog ( and @RosMcM’s ( blog points about lower ability students stand, as long as the student sits English Literature and passes English and maths.
  4. For student capable of gaining Cs it appears that it is better to get a student worse grades but have no gaps in the ebacc part of the basket in their best 8, than a kids better grades but a gap (ie one gets mixture of Cs and Ds but 3 grades in ebacc does better than a student with better grades but missing an ebacc).
  5. One further question.  With the matrix for progress being set by the exam results last year, will there be analysis on how much the current best eight measure is skewed by BTECs? –  the benchmark will be based on the 2016 exams so the possibility of BTECs taken last year skewing the average is removed

What have i learnt?

  1. This is undoubtedly a better measure than 5A*toC including (or as @timleunig put it “the 5 Cs measure”)
  2. As a school it is now vital that you get as many students as possible to have their basket of courses right.  For us we will be checking post options that every student capable has three ebacc subjects to avoid any gaps.  But, apparently the straight jacket of the old ebacc has gone, as triple scientists have their three (although question marks remain on how much double science will count…or a normal curriculum choice in schools will require only one more ebacc to ensure this part of the basket has no gaps.
  3. I have concerns about the 30% of our current Year 9s that don’t have KS2 data
  4. It will be risky to only offer 8 subjects, but the new examination system may force this as the amount of end of Year 11 exams is starting to look daunting for students
  5. I am still very nervous about the dead hand of politicians getting to this
  6. The move to pariety of English Literature and Language is very welcome – especially as the better score will count for the double as English (so a student does better in Literature that becomes the english grade that counts as part of the accountability and the language grade will count in the ‘other three’.

EDIT – @timleunig contacted me to express that the grade scores listed below had not been decided.  But, they are looking at possible replacements for the 58 for an A* and so on.  I would suggest mirroring the 1-9 that has been put forward by OFQUAL. (Update – so it was 1 -9 then!)

EDIT – I removed the calculations as they are clearly wrong!



15 thoughts on “New Accountability – thoughts and questions

  1. I’m still really interested in how they go about setting the target for the expected total best 8 points.

    The current system sees students KS2 results used to set a “target” APS based on what other similar students got nationally previous years. This target exists whatever combination of subjects a student takes – whether they be all GCSEs,or a mix of GCSEs and other courses such as BTECs.

    The problem with this is that the KS2 APS -KS4 progression for different qualification types shows that some are “easier” than others. For example progression from KS2 APS -GCSE science is significantly lower than progression from KS2 APS – BTEC science yet a grade C GCSE and a pass at BTEC are equivalent.

    It’s even true within a qualification type. Progression from KS2 APS to GCSEs in MFL are much lower than KS2 APS to GCSEs in Business Studies.

    If you were wanting to “game” the system you’d push students towards the ‘easier’ qualifications/subjects as it would appear easier for students to reach their target if you did so.

    The new accountability model has this flaw embedded too BUT they did say that they were looking at whether the model could be adapted so that the “target” score for a student was actually based on the subjects they took.

    Ie. If student X took Maths, English Lang, English Lit, Double science, History, Art, Economics their target points score might be 402 (using current points methodologies) whereas if they took Maths, English Lang, English Lit, BTEC Art, BTEC Business, Double science, Art, Economics then their target would be 418 reflecting the fact that they have taken subjects with “easier” progression from KS2 APS – GCSE.

    The way they’d do this would be to generate KS2 APS – KS4 progression tables for each different subject (for each different student) and then you’d add together the 8 subjects that the student was taking to find their end of KS4 target.

    Whilst this sounds difficult the data already exists – it is the FFT chance tables (using the FFT A data).

    The beauty, if they can make this work, is that it makes “options decisions” neutral for the school. The only thing that matters is which courses are right for the student – no gaming strategies exist that work for the school but against the student best interest.

    This is the model that is currently proposed for the KS5 accountability measures too.

    If they can make that work it’s pretty much perfect! The only flaw is what to do with students who take fewer than 8 subjects? What data is used to fill their gaps?

    1. Hi Steve

      The gaps are quite interesting. Firstly, for most of the students there shouldn’t be any gaps. If the predicted outcome from an ebacc subject is an G then the subjects needs to make sure that student gets at least a G. It stops the desire of some subjects to only gain ‘academic’ pupils (History is a good example of this). Secondly there is enough choice in the other 3 to bump up the score by gaining a Pass or higher, meaning that you should get positive VA for your lowest ability. Tim also made the point that getting one student 4 Es would be better than getting 8 Gs (4×3 = 12 points vs 8×1). Of course this has to include maths and an English. I also think your example Of student x and y both equal to 7 subjects being counted as the second student hasn’t filled their ebacc basket. Simply, as I stated you must not allow any student to not have the 3 ebacc subjects in their basket no matter how you fill it.

  2. Thanks for these comments – and for your blog on this issue. It’s really helpful to debate the detail as well as the general principles.

    To be honest I didn’t pick my 2 examples with the best 8 ‘basket’ in mind – although it would have been better if I had!

    Looking at our cohort around 85-90% meet the full basket of qualifications already without any push towards E-Bacc subjects. We made a willful decision not to push the full E-Bacc at all (other than to flag it’s existence to parents) but have seen a steady increase in take-up of those subjects as a result.

    Tim’s comments about low ability students getting 4 E’s is better than 8G’s works for Attainment8 and Progress8 if the ‘target score’ (used to calculate the progress from KS2-KS4) is done irrespective of the subjects taken (like the current KS4 APS targets). This is a real positive in terms of matching the curriculum to the needs of these students.

    I, however, prefer the idea they are looking at for calculating the ‘target score’ for KS4 based on the actual subjects a student is taking – this takes away the opportunity for schools to ‘game’ the calculations by pushing students for ‘easier’ GCSEs (those which have more favorable progression from KS2-GCSE) or ‘easier’ qualifications (perhaps BTECs or Cambridge Nationals?)

    If a student had a target of 402 points (current points system) then it would appear there are certain combinations of subjects which make it easier to hit that benchmark and schools could push students towards those subjects for the wrong reasons.

    If students take a full basket of subjects (En, Ma, 3 E-bacc + 3 others) then all is well – if not then the target stays the same but it’s unlikely that the student hits their target points as they only have, say, 7 subjects being counted. This is in line with the Government’s wish to push more towards E-Bacc subjects.

    If they go for setting the ‘target score’ using the actual subjects students take then calculating the target score for a student with a full basket of 8 qualifications is straight forward.

    My area of wonder is how they deal with setting the target score for students who do not have a full basket of qualifications (whether from doing less than 8 subjects or as a result of not doing enough E-Bacc subjects).

    I’m not sure how well I’ve explained that so I’ll give an example:

    Steve gains a level 4.9 in KS2 Maths and 5.3 in KS2 English (decimalised as they already do in their KS2-4 calculations).

    If they go for the more straight forward (but arguably ‘gamable’) method of calculating the ‘target points total’ this may come out at a target of 418 irrespective of whether Steve takes ‘harder’ or ‘easier’ subjects/qualifications.

    If they go for setting it on a subject by subject basis Steve’s target will be based on which subjects he studies (and how equally able students have done in those same subjects nationally).

    Steve may do (and have subject specific target points based on his KS2 results):

    English Lang 52.3
    English Lit 52.8
    Maths 49.8
    Double science 51.4 x 2
    Geography 52.5
    Art 51.9
    Business Studies 52.9

    Giving a total target score of 415 points.

    But what if Steve had, instead of Geography, taken Textiles? Or perhaps doesn’t do Textiles and just does 7 subjects for some reason?He then doesn’t have a complete basket of 8 subjects (he only has 2 E-Bacc subjects) – how would we then calculate his target points score? We only have 7 individual subject targets to add up! What would his target be now?

    If we use this total of 7 subjects as his target (and just use his 7 subjects to see how he progresses) then the whole basket of 8 becomes meaningless. Do they fill the ‘blank’ slot with a dummy subject that he doesn’t take? Or perhaps the target for textiles that he takes (but just not count his result in the calculations)?

    I hope they can find a solution because, if they do, subject specific ‘targets’ it seems to be the fairest way.


    1. I think you need to think about the APS not the tps. total score is divided by 10 to give you your APS. So if you only do 7 subjects that count (doubling maths and English giving you 9 in total) the total will still be divided by 10.

      It is the APS that gives the value added. So score 29 at KS2 national progression from that is 8Cs (giving you a total points score of 50 points as en/Ma double) APS = 5. Lose one from your basket (total points score = 45 ALS 4.5 you are 0.5 below where you should have been with this student). Remembering that being below 0.5 mean you don’t meet your floor target

      1. But the DfE are looking at possibly calculating the APS points score based on the the subjects you take.

        My particular area of interest is how they will, if this is the case, calculate the target TPS (or APS).

        I understand that the proposal to use the actual combination of subjects taken means it won’t be as simple as saying 29 at KS2 equates to TPS of 50 at KS4 – if you take one particular combination of subjects your TPS (target) may be 49; take a different combination of subjects and your TPS (target) may be 52 (or 4.9 and 5.2 respectively if we use APS).

        I understand that if a student takes fewer than 8 subjects in the basket then they’ll be more likely to miss their APS target.

        I am just unclear as to how the TPS (or APS) target will be set if a student only takes 7 subjects.

        If they do 8 (allowable) subjects then the APS target will be calculated by adding the 8 individual subject target target points scores; adding in maths and English again (as they count double) and then dividing by 10.

        If they do 7 (allowable) subjects then the APS target will be calculated by adding the 7 allowable individual subject target points scores; adding Maths and English again and then what? Dividing by 9 makes the notion of ‘basket’ irrelevant; dividing by 10 makes no sense as that makes the APS target lower. They need to pit an 8th subject target in there somehow to make things fair – but no 8th subject is being taken!

        Agree very much in principle but, previous accountability measures, have failed because not enough consideration has been given to the detail and how it might play out in schools. I’m just keen that every possible angle is looked at to ensure that we don’t end up with some more unitneded consequences.

      2. My understanding is that as a school, as far is possible, don’t have children doing less than 8. We decided to go for a 9 GCSE model and will ensure no student drops subjects so they end up with less than 8. A poor result in one subject will still give some points towards the APS.

    2. However I do agree with your point about some weighting in the subjects…but then the ‘harder subject’ debate comes back. From memory CEM found RM to be one of the hardest subjects and history and geography were a long way down the list

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s