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

Exam marks for certain subject are unreliable

Discussion in 'Education news' started by Lalex123, Dec 2, 2018.

  1. Lalex123

    Lalex123 Established commenter

    agathamorse likes this.
  2. Toomuchtooyoung

    Toomuchtooyoung Occasional commenter

    Had a gcse Biology reviewed, 3 marks were found and the grade upped to a 7. We had to pay upfront and were lucky we could afford it, think it was around £80. Another friend paid for a priority review of A level physics, the marks were found but too late and a university place lost. Of 5 reviews families and friends put in this year for gcse and A level only one didn’t result in an increased grade.
    agathamorse likes this.
  3. aypi

    aypi Senior commenter

    When in school I got C's when I was marked by the class teacher and A's when marked by the exam board. A friend in the top set gave his essay to another in the 7th set. The top setter got 90% the 7th setter 36%. Same essay, different prejudice.
    Shedman and agathamorse like this.
  4. Catgirl1964

    Catgirl1964 Senior commenter

    Having a difference in the accuracy of marking in subjective subjects such as English is understandable but not to the extent mentioned. Maybe a solution could be that several markers check subjective subjects' papers to eliminate controversial results. Just a thought.
  5. ridleyrumpus

    ridleyrumpus Star commenter

    Whilst not defending this, there can be other factors at play. Sometimes, esp with lower set pupils, it can be impossible to read a students writing.

    I have students who cannot read their OWN writing....
    drek, colpee and border_walker like this.
  6. aypi

    aypi Senior commenter

    The lower set writing was clear.
  7. MrMedia

    MrMedia Star commenter

    The grade boundaries are too narrow. In the more subjective subjects you should jump in double grade boundaries and there should be 'gaps' between grades from the 8 to the 2.

    At GCSE whether they get a 7 or an 8 is quite immaterial to the standard of knowledge. The contextual variables affecting the outcomes in the shape of 'noise' are greater than the current grade boundaries.

    Bigger boundaries and gaps.
    82+ outstanding
    62-78 merit
    42-58 pass
    Below 38 fail
    drek and Shedman like this.
  8. FrankWolley

    FrankWolley Star commenter

    Hmm... I spent over 30 years teaching History. It isn't a 'subjective subject', in the sense that marks (or grades, or levels) are simply awarded according how how the marker 'feels' (Definition of subjective: 'based on or influenced by personal feelings, tastes, or opinions'*.). So please give the thousands (tens of thousands?) of History teachers out there for knowing perhaps a tad more than you appear to about how they grade work in their subject!:mad:

    damia69 likes this.
  9. blazer

    blazer Star commenter

    They can't find markers now. Where are the extra markers/hours going to come from and who is going to pay?
    JohnJCazorla likes this.
  10. adam_nichol

    adam_nichol Occasional commenter

    MrMedia is rather right. I'm a former teacher and current examiner.
    Questions with a set right/wrong criteria are best off machine marked really.
    Those that are subjective tend to be impacted by a squish in marking. Examiners qualify and are checked upon every so often for accuracy (seeding or random sampling normally). It is risky to assign top or bottom level marks in case you later get a better or worse one with no way to differentiate that. Exam boards know this and are taking measures to tackle, but you end up fighting a potent psychological process. The marks for all the papers are then within a small range and individual differences over a range of questions can add up to more than a grade boundary.

    An alternate is to ditch the concept of marking to a set criteria, and instead use comparative judgements - examiner sees 2 scripts and ranks one as better than the other; then sees 2 more and so on. Statisticians will be able to advise on how many need to be ranked to get an accurate placement of all scripts on a distribution curve, from which you then draw grade boundaries from standard deviations from the median.

    Australia have been exploring AI marking. As yet, it's not quite there (and would benefit from typed exams - but that's a while other rant of mine). However, AI is coming....
  11. Lalex123

    Lalex123 Established commenter

    I merely quoted from the article. I’m guessing the journalist researched the largest gaps in marks across a number of subjects and did a lot of research. This could also be because the subject has a heavy literacy focus.

    I’m not quite sure why this would make you angry - subjectivity is not something that is an exact science... maths however cannot be subjective so the marks are always correct... or not!
  12. FrankWolley

    FrankWolley Star commenter

    I think you misunderstand both me, and the examining process, as much as the journalist clearly does.

    I could have added that I was a GCSE examiner (16 years) and A Level examiner (4 years), and - of course - also marked literally thousands of internal exams over 30+ years. So I think I know a little bit about the process of assessment of history scripts. And if it was 'subjective', then I suspect I (& my colleagues) would have clocked that decades ago. And would have done something about it.

    NB For future reference, if you want to copy someone else's comments, do it as a quote, otherwise we'll assume it is what YOU think... ;)

    PS I'm no Mathematician, but I'm petty certain I've heard a few Maths teachers over the years deny that their subject is totally objective, esp. the higher you move up it, as credit is given more & more for the working and methods used, and not just for the (right or wrong) answer.
    damia69 likes this.
  13. border_walker

    border_walker Lead commenter

    I seem to recall DT coursework being marked in a similar way. Pairs were simply ranked by 2 different teachers. from this a computer created an order.
  14. drek

    drek Star commenter

    And perhaps the teachers of the two sets found about about the cheating and downgraded the essay because of that?

Share This Page