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

More cracks appear in the system

Discussion in 'Education news' started by Mangleworzle, Oct 6, 2015.

  1. Mangleworzle

    Mangleworzle Star commenter


    "Private school heads condemn exam marking"

    Well not so much appear as it becoming more difficult to keep looking the other way.

    "Last year more than 400,000 exam papers were challenged, with more than 77,000 grades being changed."

    That's 19% of those questioned. My school used to request remarks only for those students who were 2-3 marks away from the next grade boundary up depending on if we could afford it, maths used to put loads of remarks in as they were given the most money to pay for better results as their results affected the school more than my subject (science).

    If the marks went down then those students were far enough into the grade for it not to drop them a grade they went from high to middle or maybe low, but still in the grade. Presumably those at the bottom end of the boundary would have around a 19% chance of going down, but of course no-one is going to question scraping into the next grade up.

    A friend of mine questioned one of her daughters gcse exams a couple of years ago, on the re-mark, the overall grade went from an E to A*, they neglected to add all the marks to the spreadsheet apparently.
  2. foxtail3

    foxtail3 Star commenter

    What's the reason? Are they having to mark papers too quickly? Are they not up to the job? Are there too many variables?

    There can be no excuse surely, for the example you quote!
  3. Vince_Ulam

    Vince_Ulam Star commenter

    It's a poor private school which does not budget grade challenges into their service.
    stmha likes this.
  4. Mangleworzle

    Mangleworzle Star commenter

    Which just underlines the problem, it is possible to buy better grades for some students.
  5. stmha

    stmha Established commenter

    I still do exam marking for OCR and their checks are stringent. Everyone checks everyone who then checks...everyone. I can only assume this is not poor marking but more a transcription error
  6. Vince_Ulam

    Vince_Ulam Star commenter

    It's a game rather than a transaction but league-wise it's proven sufficiently profitable over a sufficient number of rounds for some schools to delude themselves that they are entitled to a particular proportion of appeals in their favour.
    Last edited: Oct 6, 2015
  7. Mangleworzle

    Mangleworzle Star commenter

    The 19% figure overall seems quite low compared to my subject where our success rate was closer to 1/3rd, then again we had less money to spend and so had to target more closely.

    However the point is that the system should be sufficiently trustworthy so as not to require paid-for remarks. I wonder when these started? and why they started, do the exam boards see them as just another income top-up opportunity?
  8. Vince_Ulam

    Vince_Ulam Star commenter

    I am sure that you do not believe there should be a relationship between your outcome and King's figure but you can see how casual comparative language could suggest that one exists. The HHC assume the relationship.

    I do not think the HHC object to the payments provided they do not feel cheated by their outcomes.
    Last edited: Oct 6, 2015
  9. Eureka!

    Eureka! Lead commenter

    It's the foundations that are rotten.

    I mean, how ridiculous is it to funnel percentages zones into grades? Begging for trouble!
  10. Vince_Ulam

    Vince_Ulam Star commenter

    Thank you, random-man-in-the-street.
  11. Mangleworzle

    Mangleworzle Star commenter

    I was assuming a little greater understanding on the part of the reader, oh well.

    Is there a point to this statement of what you reckon?
  12. MrMedia

    MrMedia Star commenter

    We are focusing on the wrong aspect.

    It did it for level descriptors, but there is very little between each grade - the factors that can affect movement from one grade to the next could be as little as hay fever, a single badly marked question, a poor night's sleep, salty food and dehydration and a whole bunch of other things. The main thing is that there is very little between grades. You can't appeal hay fever, but you can an examiner.

    There being very close grades would not really matter if the difference between one grade and another were not HIGH STAKES. Headteachers' jobs, people's pay rises, places at FE, sixth form and university, floor targets etc are all loaded onto this ludicrously thin gossamer of a grade descriptor.

    Just like schools and Good/Outstanding and Good/RI the stakes on the border are artificially high.

    Schools, students, all of them. There needs to be a big enough difference between one grade and the other.
  13. Vince_Ulam

    Vince_Ulam Star commenter

    You would often be right to do so but why did you make the comparison at all? I can't assume your motive but in my experience many other people assume that proportions in the data immediate to them should communicate through all subsets of the set in which their data exists.

    Yes. It states what I reckon.
  14. Mangleworzle

    Mangleworzle Star commenter

    Eh? My point was that it implies that many remarks are probably more speculative than ours, hence the proportion is different.

    A greater number of speculative remarks will cost more but ultimately deliver more improved grades, we have a system where it possible that schools and hence all of the things that Mr. Media mentions can be improved by payments to exam boards if you have enough in the kitty. A broken system.
  15. Vince_Ulam

    Vince_Ulam Star commenter

    Your suspicion of speculation itself assumes a regularity integral to the system. There is speculation because, as I have said, some schools assume a regularity from a temporary winning streak.

    Well, it will ensure that papers are more rigorously graded which is not necessarily going to result in an increase in higher grades. There will ultimately be additional costs. Were I the HHC I should have kept quiet.

    That the HHC complain increases my confidence in the system. Some gaming is inevitable but the House always wins.
  16. Mangleworzle

    Mangleworzle Star commenter

    What a peculiar conclusion. You think a system that is so unreliable is one to have confidence in?

    If it was reliable the change in grades would be much less than 19%. There is much noise affecting performance on the day but marking of the script produced on the day should be more consistent than this.
  17. Vince_Ulam

    Vince_Ulam Star commenter

    I assume that you had a measure of confidence in the system such that you would not challenge A*s nor challenge expected Gs. The contrary would be peculiar.

    According to which measure? If we assume incompetence then we cannot assume that the favourable new grades awarded within this 19% were reached any more rigorously than they previously were.

    How consistent would you like them? Pick a number.
  18. Eureka!

    Eureka! Lead commenter

    Apart from the rationale for needing this, you need to realise that consistent labels in artefacts can only be constructed, not measured.

    I can buy different sizes of screw, but any attempt I make to classify the weather into types is going to be subject to classification overlap.
  19. Vince_Ulam

    Vince_Ulam Star commenter

    That you cannot measure them only means that you do not have the education, the training and the experience to measure them.
    Last edited: Oct 6, 2015
  20. Mangleworzle

    Mangleworzle Star commenter

    I'm not saying it should (or could) be expected to be perfect. Over the years I have seen the challenge of marking go from an extremely rare event to a yearly ritual, from something that only particularly pushy parents would do to something budgeted for by the school in advance. That wouldn't have happened if it wasn't worth it. I still have my suspicions that it may also be another useful income stream by the boards at an otherwise quiet time of year.

    Better than 19% - the lower the better.

    If a school is this far out in estimating (and that's estimating) grades and levels, there is a whole world of poopy stuff to go through. Organizations that tout themselves as being approved to make such absolute measures should be far, far better at internal consistency. After all, budgets, schools, careers, university places and more depend on it.

    We're not even talking about year to year consistency here but internal consistency following the same mark scheme for the same exam within a few weeks.

Share This Page