Turn on thread page Beta
    Offline

    18
    ReputationRep:
    (Original post by 13 1 20 8 42)
    I lost it when the eigenvalues were really weird...couldn't believe I'd done it right and kept looking for a mistake then I just left or barely bothered with half the proofs properly lol. Dropped like a dozen marks
    Yep they were weird haha. I thought I had 100ums but I just missed out.. 1 in first question then like another 6 in last part and na 3 marks in the distance part. Dropped a 1/2 in the reduction proof for not showing what I_0 or I_1 was.
    Offline

    10
    ReputationRep:
    (Original post by Euclidean)
    For Spearman's RCC, how are 'Tied Ranks' supposed to be dealt with?

    Also, is it fine to write:

    H_0: There is no correlation between X and Y

    Rather than:

    H_0: \rho = 0
    bump
    Offline

    12
    ReputationRep:
    (Original post by Euclidean)
    bump
    You should put it in terms of Roh, the mark schemes are really strict about this :s
    Offline

    10
    ReputationRep:
    (Original post by Ayman!)
    You should put it in terms of Roh, the mark schemes are really strict about this :s
    Thanks again
    Offline

    2
    ReputationRep:
    (Original post by Euclidean)
    bump
    (Original post by Ayman!)
    You should put it in terms of Roh, the mark schemes are really strict about this :s
    I always do it as H0: p=0 (there is no correlation between x and y)
    Offline

    2
    ReputationRep:
    (Original post by Ayman!)
    You should put it in terms of Roh, the mark schemes are really strict about this :s
    what is roh the greek latter xD
    Offline

    12
    ReputationRep:
    (Original post by L'Evil Wolf)
    what is roh the greek latter xD
    \rho - this thingamabob

    (Original post by Euclidean)
    Thanks again
    No worries, man!

    Does anyone know what we have to do when we have tied ranks? Is it just "average ranks and use PMCC"?
    Offline

    7
    ReputationRep:
    (Original post by Ayman!)
    \rho - this thingamabob



    No worries, man!

    Does anyone know what we have to do when we have tied ranks? Is it just "average ranks and use PMCC"?
    Yes exactly what you said. If you go on the Solomon papers (B i think) there's a question like that that asks "what would you do if the price for B and F (or some other letters) was the same" and the answer is to use average ranks then use PMCC
    Offline

    22
    ReputationRep:
    (Original post by Ayman!)
    \rho - this thingamabob
    It's rho, not roh.
    Offline

    12
    ReputationRep:
    (Original post by Zacken)
    It's rho, not roh.
    Realised when I tex'd it...
    Offline

    10
    ReputationRep:
    (Original post by Ayman!)
    Does anyone know what we have to do when we have tied ranks? Is it just "average ranks and use PMCC"?
    (Original post by coolguy123456)
    Yes exactly what you said. If you go on the Solomon papers (B i think) there's a question like that that asks "what would you do if the price for B and F (or some other letters) was the same" and the answer is to use average ranks then use PMCC
    What does it mean by 'average ranks' and then use product moment correlation coefficient? (I assume that's PMCC)
    Offline

    7
    ReputationRep:
    (Original post by Euclidean)
    What does it mean by 'average ranks' and then use product moment correlation coefficient? (I assume that's PMCC)
    If you have B and rank 6 and F at rank 7 and the question if i remember correctly it said the price of B and F is now the same so you'd rank them 6.5 each and then figure out Sxx and Syy and Sxy and then use PMCC.
    Offline

    10
    ReputationRep:
    (Original post by coolguy123456)
    If you have B and rank 6 and F at rank 7 and the question if i remember correctly it said the price of B and F is now the same so you'd rank them 6.5 each and then figure out Sxx and Syy and Sxy and then use PMCC.
    Ah thanks a lot!
    Offline

    2
    ReputationRep:
    Name:  s3.JPG
Views: 114
Size:  45.0 KB
    can someone explain part d)? its from the June 15 paper
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by Achint10)
    Name:  s3.JPG
Views: 114
Size:  45.0 KB
    can someone explain part d)? its from the June 15 paper
    I can't remember what the answer is or how to get it, but it's the most stupid question in S3 that I have ever seen.
    Offline

    12
    ReputationRep:
    (Original post by Achint10)
    Name:  s3.JPG
Views: 114
Size:  45.0 KB
    can someone explain part d)? its from the June 15 paper
    Zacken just explained this to me a few hours back.

    (Original post by Zacken)
    It's saying that you need to say two things: 1. In the individual strata, the variances are small. 2. In the combined thing, the variances are large since the difference of means are large meaning the overall variances from the overall mean are large.
    Offline

    22
    ReputationRep:
    (Original post by Achint10)
    Name:  s3.JPG
Views: 114
Size:  45.0 KB
    can someone explain part d)? its from the June 15 paper
    If you consider both the children and the staff together, the variances from the mean will be quite large since children and adults have (obviously) vastly different weights.

    If you consider the children and staff separately, the variances in each will be much smaller.

    Since standard error is proportional to the square root of variance, you're there.

    (Original post by SeanFM)
    I can't remember what the answer is or how to get it, but it's the most stupid question in S3 that I have ever seen.
    How come? It's actually a valid question that isn't stupid sampling crap.
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by Zacken)


    How come? It's actually a valid question that isn't stupid sampling crap.
    You're almost saying nothing new at all, and it's not quite clear what kind of comment it wants you to make, though I suppose the 'give a reason for any differences' kind of gives it away a bit. It's really not something you'd want to be asked in the pressure of an exam - it'd just be 'oops, there's 2 marks gone'.
    Offline

    2
    ReputationRep:
    (Original post by Ayman!)
    \rho - this thingamabob



    No worries, man!

    Does anyone know what we have to do when we have tied ranks? Is it just "average ranks and use PMCC"?
    Yes you give the average. Although most questions will tell you what to do, I have never came across an edexcel question asking what you mentioned.

    They generally tell you which one to choose.

    However, if tied at 6 (2 of them) and there is obviously no 7th place, give ranks of 6.5 and use PMCC i believe

    (Original post by Euclidean)
    What does it mean by 'average ranks' and then use product moment correlation coefficient? (I assume that's PMCC)
    Offline

    10
    ReputationRep:
    Is there a penalty for arranging the ranks in reverse order in SRCC?

    In the below question I ranked the average attendance in ascending order (C as 1) rather than descending order (B as 1). The value of the coefficient is negative as a result. I don't think the question specified very well their intentions? (It was the specimen paper though)

 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: June 29, 2016
Poll
Cats or dogs?

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.