Turn on thread page Beta
    Offline

    0
    ReputationRep:
    (Original post by Tallon)
    I would have thought it was implied because the question was in the context of a test and you surely only get integer marks in a test.

    We really need to ask somebody who actually knows...
    That would be the best. I wouldn't have thought it would be more than a mark, and you can get 1/2 marks in tests, especially internal ones. That said I doubt anyone has ever got 54.322/72 in a test for example. Could try and find a past paper.
    • PS Helper
    • Thread Starter
    Offline

    4
    ReputationRep:
    PS Helper
    (Original post by Chriz M)
    That would be the best. I wouldn't have thought it would be more than a mark, and you can get 1/2 marks in tests, especially internal ones. That said I doubt anyone has ever got 54.322/72 in a test for example. Could try and find a past paper.

    ugh gross, if you got half marks you'd need a 0.25 continuity correction. I don't know.

    I imagine it's like significant figures where they only mark you down for it once? Hope so anyway. Maybe they'll realise that their question was too ambigeious and mark it both ways.
    Offline

    0
    ReputationRep:
    I'm pretty sure you only need to use a continuity correction when approximating a discrete distribution, i.e. poisson or binomial, with the normal distribution.
    • PS Helper
    • Thread Starter
    Offline

    4
    ReputationRep:
    PS Helper
    (Original post by DaGianni)
    I'm pretty sure you only need to use a continuity correction when approximating a discrete distribution, i.e. poisson or binomial, with the normal distribution.

    marks in a test is a continous measurement then?
    Offline

    0
    ReputationRep:
    (Original post by Chriz M)
    That would be the best. I wouldn't have thought it would be more than a mark, and you can get 1/2 marks in tests, especially internal ones. That said I doubt anyone has ever got 54.322/72 in a test for example. Could try and find a past paper.
    To quote my self, but I did have a look, and according to MEI's own resources. (First Chapter of Normal bit, in Notes & Examples), they use tests as an example and do no use a continuity correction. I suspect, it is fine as the normal distribution is continuous. Otherwise we would only lose a couple of marks, could be interesting to see the mark scheme, I'll ask my teacher tomorrow.
    Offline

    0
    ReputationRep:
    (Original post by Straightpath)
    The answer has to be 61, as a student can only get either 60 or 61 in the test, and if they get 60 then thats not in the top 10%, 61 is.

    Here's the paper:
    Can some1 who knows what they're doing put the answers up using the attachments to the paper in this person's post?
    Offline

    0
    ReputationRep:
    (Original post by Ciara222)
    Can some1 who knows what they're doing put the answers up using the attachments to the paper in this person's post?
    I think the issue is with that question no-one seems to be certain of the answer.
    Offline

    0
    ReputationRep:
    (Original post by Tallon)
    Also, for the benefit of my friend who's asking me, what did you do for the question about the probability of getting a score of 111? I did P(110.5<x<111.5) using the new mean and SD they give you. Not entirely sure if that's right though.

    As on the bottom side of the inequality there is an inclusive sections. Larger than 110.5 inclusive. The normal distribution cannot do inclusive so you have to do a correction and do 110<x<111.5 that's what I did. Oh well did it early anyways. Going for the A*. Hopefully get >90% otherwise just have another go in June.

    Thank god we didnt have any waffly question like when they give you the tables on the last question and ask you to comment on each of the different things(e.g. species, musical preference) blah blah blah.
    Offline

    0
    ReputationRep:
    (Original post by Tallon)
    marks in a test is a continous measurement then?
    You would think not, however the question tells us that the marks are distributed normally (as opposed to being actually distributed with a discrete distribution and then merely approximated by a normal distribution), so I just took this literally and assumed the marks were continuous. I'll be interested to read the examiners' comments for this particular question once the mark scheme becomes available for viewing online, it certainly seems to be the source of much confusion.
    Offline

    0
    ReputationRep:
    (Original post by RenoC)
    As on the bottom side of the inequality there is an inclusive sections. Larger than 110.5 inclusive. The normal distribution cannot do inclusive so you have to do a correction and do 110<x<111.5 that's what I did. Oh well did it early anyways. Going for the A*. Hopefully get >90% otherwise just have another go in June.

    Thank god we didnt have any waffly question like when they give you the tables on the last question and ask you to comment on each of the different things(e.g. species, musical preference) blah blah blah.
    I did that too, especially as that part of the question specifically states the score is rounded to the nearest integer, so perhaps if they said this for that part of the question, rather than putting it in at the start of the question, suggest it could have continuous raw marks.
    Offline

    8
    ReputationRep:
    (Original post by DaGianni)
    You would think not, however the question tells us that the marks are distributed normally (as opposed to being actually distributed with a discrete distribution and then merely approximated by a normal distribution), so I just took this literally and assumed the marks were continuous. I'll be interested to read the examiners' comments for this particular question once the mark scheme becomes available for viewing online, it certainly seems to be the source of much confusion.
    The best outcome for everyone would be that it's continuous, but the mark scheme says to ignore any attempt at rounding to integers.
    Offline

    0
    ReputationRep:
    (Original post by Game_boy)
    The best outcome for everyone would be that it's continuous, but the mark scheme says to ignore any attempt at rounding to integers.
    Which mark scheme?
    • PS Helper
    • Thread Starter
    Offline

    4
    ReputationRep:
    PS Helper
    (Original post by RenoC)
    As on the bottom side of the inequality there is an inclusive sections. Larger than 110.5 inclusive. The normal distribution cannot do inclusive so you have to do a correction and do 110<x<111.5 that's what I did. Oh well did it early anyways. Going for the A*. Hopefully get >90% otherwise just have another go in June.

    Thank god we didnt have any waffly question like when they give you the tables on the last question and ask you to comment on each of the different things(e.g. species, musical preference) blah blah blah.
    Sorry?
    p(x=111)
    you did P(110<x<111.5)? Why didn't you do P(110.5<x<111.5)?
    I don't quite understand.
    Offline

    8
    ReputationRep:
    (Original post by Chriz M)
    Which mark scheme?
    No; I mean the best outcome would be if the mark scheme said [...].

    @Tallon

    Yeah I'm not getting this 110 thing. You're already accounting for discrete -> continuous by doing 110.5 to 111.5. I think others are getting confused with it looking a bit like a continuity correction.
    Offline

    0
    ReputationRep:
    (Original post by Tallon)
    Sorry?
    p(x=111)
    you did P(110<x<111.5)? Why didn't you do P(110.5<x<111.5)?
    I don't quite understand.
    I did that because to get 111 marks you have to get 110.5<=x<111.5
    You can get 110.5 and the normal distribution does not work with equality. it is either less than or more than that the normal works with. You must do continuity corrections. Anyways I am no teacher so don't ake my absolute word for it.
    • PS Helper
    • Thread Starter
    Offline

    4
    ReputationRep:
    PS Helper
    I skipped all my maths lessons today so I didn't ask any of my teachers for Q3 answers

    did anyone?
    Offline

    8
    ReputationRep:
    (Original post by Tallon)
    I skipped all my maths lessons today so I didn't ask any of my teachers for Q3 answers

    did anyone?
    My teacher said integer, because it's a test. But that's no better of a guess than any of us can do without the mark scheme; the examiner could justify it either way.
    Offline

    0
    ReputationRep:
    Mine said you only need a continuity correction if modelling the poission or binomial, or if like the scaling question, specifically requires rounding.

    The continuity correction is required for rounding. The raw marks aren't rounded. If you get 5.5 marks then that is your raw mark. If you got 50 raw marks, then you won't have 49.5-50.5 marks you would have 50. Again we can't be certain until the markscheme, but I don't think it is needed.
    • PS Helper
    • Thread Starter
    Offline

    4
    ReputationRep:
    PS Helper
    (Original post by Chriz M)
    Mine said you only need a continuity correction if modelling the poission or binomial, or if like the scaling question, specifically requires rounding.

    The continuity correction is required for rounding. The raw marks aren't rounded. If you get 5.5 marks then that is your raw mark. If you got 50 raw marks, then you won't have 49.5-50.5 marks you would have 50. Again we can't be certain until the markscheme, but I don't think it is needed.

    sucks to argue about it but:

    if you get half marks (in your example, 5.5) the continuity correction for P(x=5.5) would be P(5.25<x<5.75), no?

    if you don't use a continuity correction, you're saying X can take any value. So you're saying there's a probability of getting 3.575837434837436839437843 marks. Clearly this is rubbish. Surely, surely, surely you need continuity corrections?
    Offline

    0
    ReputationRep:
    (Original post by Tallon)
    sucks to argue about it but:

    if you get half marks (in your example, 5.5) the continuity correction for P(x=5.5) would be P(5.25<x<5.75), no?

    if you don't use a continuity correction, you're saying X can take any value. So you're saying there's a probability of getting 3.575837434837436839437843 marks. Clearly this is rubbish. Surely, surely, surely you need continuity corrections?
    I see the point and would certainly be true if approximating binomial or poisson, but this wasn't an approximation. A problem is if you try find the probability of getting 50 marks or above for example, then you aren't finding that if you include 49.5 and above, like if you took a continuity correction.

    I doubt we will agree, I didn't use one, I suspect you did, so neither of us want to be wrong.
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: March 12, 2010

University open days

  • University of Exeter
    Undergraduate Open Days - Exeter Campus Undergraduate
    Wed, 24 Oct '18
  • University of Bradford
    Faculty of Health Studies Postgraduate
    Wed, 24 Oct '18
  • Northumbria University
    All faculties Undergraduate
    Wed, 24 Oct '18
Poll
Who is most responsible for your success at university

The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

Write a reply...
Reply
Hide
Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.