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

Cohorts less than 10 got CAG grades?

Discussion in 'Mathematics' started by Danio, Aug 13, 2020.

  1. Danio

    Danio New commenter

    Is this true? If cohort was less than 10, then no standardisation was done, and they were awarded their teacher assed grades? Seems incredibly unfair to me and biased towards smaller sixth forms/colleges.

    Alevel maths - 15/19 downgraded from CAG

    Alevel FM - all got CAG (5 students)

    Mathematical Studies - all got CAG (6 students)
  2. install

    install Star commenter

    I thought it was cohorts less than 5. It’s because the statistical analysis wouldn’t work on small cohorts so they have to rely on the Teacher Assessment being accurate.

    Basically if you were in a small cohort or home schooled you might be onto a winner. Which is very unfair. Imagine the home schooled students (not in a large cohort) sending off top grades in their Teacher Assessments and finding they had all been approved.
  3. Danio

    Danio New commenter

    Ok, I am not sure exact numbers for those smaller groups, I will check. But seems incredibly unfair that if you had a small class all got the grades teachers estimated.

    The bias here is probably towards private schools as well.
  4. pi r squared

    pi r squared Occasional commenter

    I understand that statistical modelling cannot work on a tiny cohort, but if you have the two facts (1) small cohorts get their CAGs and (2) overall standards are to remain broadly the same, then if the small cohort CAGs end up being on the generous side, there has to be an "overcompensation" for the larger cohorts in order to maintain standards. Which just screws over those centres with larger cohorts, who in many ways have a harder job in the first place because their classes are bigger! Double whammy. Thanks for that (!)
    Catgirl1964 likes this.
  5. jpayne27

    jpayne27 New commenter

    I totally agree with your point, it means state pupil generally get screwed.
    I also withdrew 5 pupils this year who were likely to get U's, a response to the U's last year rather than due to Covid. I guess the modelling assumes this did not happen and therefore applies some U's to pupils in place of those not entered. Bizarrely twice as many as last year, despite the fact we upped the entry requirements for this cohort so they were brighter anyway.
    It seems the government have grasped the importance of maintaining big data whilst ignoring the fact it produces winners and losers within the system which is hard to justify. It is also not a surprise only the losers are making a fuss.
    Catgirl1964 likes this.
  6. Danio

    Danio New commenter

    It is good to see that without me saying too much to bias your responses, but you have confirmed some of my thoughts.

    The main issue for me, is that a reported difference in the rise in top grades for independent schools compared to state school is surely affected by this cohort size rule.

    It is a major flaw without this, I saw a.tweet on twitter about how a student got an A in Further Maths but a C in Maths. Does that seem fair. You could change his to A A, but is that fair on any other student who was downgraded to a C and did not take further maths?
  7. briceanus

    briceanus New commenter

    Small cohorts make downgrading statistically troublesome is nonsense. If 40% of the general population got downgraded, then 40% of small cohorts should get downgraded. Our Latin class have now come out with the best results they've ever had, and best in school. 4 A stars and an A. Our maths cohort has seen 52% downgraded, some by 2 grades, no explanations. English lit have been shafted, all but 3 of them have been downgraded, including a student who won a very prestigious national English Lit award !
  8. Danio

    Danio New commenter

    I feel this should be a headline, but no news outlet is talking about it.
    LiamD and install like this.
  9. install

    install Star commenter

    Agree they should be. It certainly isn’t fair that a student can get a higher grade simply because they are in a smaller cohort.

    The TES did report it for the GCSEs:

    Exclusive: Teacher grades ignored in most GCSE results
    By Catherine Lough on 07 August 2020

    ..Most of this summer's grades will be based on statistical modelling of pupil prior attainment, schools’ historic performance data and the rank order of pupils in every subject submitted by schools to exam boards, sources say.

    The exception will be where there are subjects with no more than five entries in a school. In these cases, pupils will be awarded their teacher-assessed grades, as the statistical modelling would be inaccurate with such a small cohort....’
    Last edited: Aug 13, 2020
  10. pi r squared

    pi r squared Occasional commenter

    Unfortunately it's even worse than that because the 40% includes those small cohorts. If you imagine every small cohort contributing 0% downgrades then the bigger cohorts are potentially contributing 60-80% downgrades in order to "average" 40%. This also ties anecdotally with conversations I've had with colleagues in other schools and within our own school, where 60-80% seems to be quite a common figure.

    On the face of it though, Maths should be one of the most consistent subjects across the nation, because even for a small private school a Maths cohort of less than 5 is pretty unusual. Yet I still fear we'll see a disparity even so...
  11. hhhh

    hhhh Star commenter

    Thought home-schooled couldn't get a grade at all? Which seems unfair-especially as so many might do this for medical reasons.
    install likes this.
  12. install

    install Star commenter

    It seems many didn’t but that some may have depending on their postcode and local school/Centre support.
  13. jcstev

    jcstev New commenter

    No - Ofqual decided that instead of doing this they would allow a degree of grade inflation, because it would be unfair to penalise students in large cohorts for the expected over generosity in small cohort CAGs (source: Ofqual interim report which gives details of the process).
  14. pi r squared

    pi r squared Occasional commenter

    That may be the stated aim but I am pretty sure that when the dust is settled, it will become apparent that larger cohorts have been disadvantaged to a greater extent than smaller. I have read through the explanation of the process in the report and on the face of it it appears robust, but with access to only half of the variables required for the formula it is very difficult to really do much with it.

    Ultimately, the results in our large cohort school were downgraded to the extent that our value added in most subjects would be markedly worse than last year, which flies in the face of my understanding of the process from the report. It is hard to conclude anything other than an overcompensation to make up for smaller cohorts' inflated CAGs.
  15. Danio

    Danio New commenter

    So it seems like there are three tiers of this

    1-5 cohort - CAGs
    6-15 cohort - CAGs taken into account
    16+ cohort - standardized and CAGs ignores

    "In blog today for FFT Education Datalab, statisticians Philip Nye and Dave Thompson point out that teacher-assessed grades were awarded in subjects with a small number of students – five or fewer – which were "typically higher" than those generated by Ofqual's moderation process.

    For cohorts of between five and 15 students, a combination of the teacher-assessed grading and statistical moderation was used."

Share This Page