Turn on thread page Beta
    Offline

    2
    ReputationRep:
    is the unbiased estimator proof in our spec?
    Offline

    1
    ReputationRep:
    (Original post by Inges)
    PEOPLE! Fire away questions! This is my revision since I've done all the past papers to the point where I remember the mark schemes !
    hey, is total error the same as total mean error? probs wont come up buh cant stop stressing LOL
    Offline

    3
    ReputationRep:
    (Original post by Krollo)
    Not bad. I got 73/75 on that one this morning, we'll smash this bro. I have a feeling it'll be an easyish standard paper like s3.


    Posted from TSR Mobile
    Your right, s3 is repetitive, they get less and less leeway in the variation of the questions the higher up a module you go!
    Offline

    2
    ReputationRep:
    In a question that involves contingency tables, I wonder if it is necessary to draw the contingency table with the values of expected frequencies in it and then the table that shows (O-E)^2/E or is it just necessary to put the expected frequencies directly in the table of (O-E)^2/E with the expected frequencies worked out directly from your calculator? Because from the mark scheme, to my understanding is that they only look at if the values are correct or not.
    I know the question is a bit silly, but I am thinking if the time for doing the first table could be saved?
    Offline

    2
    ReputationRep:
    (Original post by AakashG)
    hey, is total error the same as total mean error? probs wont come up buh cant stop stressing LOL
    Hi, can I ask which paper total mean error has come up in? I've never heard of that before lol, don't remember it from the book. Thanks.
    Offline

    3
    ReputationRep:
    (Original post by AakashG)
    hey, is total error the same as total mean error? probs wont come up buh cant stop stressing LOL
    Which paper did they use those words on may I ask?

    They usually use "Standard error" or "standard error of the mean" so seems unusual to me!
    Offline

    3
    ReputationRep:
    (Original post by paradoxequation)
    Hi, can I ask which paper total mean error has come up in? I've never heard of that before lol, don't remember it from the book. Thanks.
    I know right!
    Offline

    3
    ReputationRep:
    (Original post by Geraer1)
    In a question that involves contingency tables, I wonder if it is necessary to draw the contingency table with the values of expected frequencies in it and then the table that shows (O-E)^2/E or is it just necessary to put the expected frequencies directly in the table of (O-E)^2/E with the expected frequencies worked out directly from your calculator? Because from the mark scheme, to my understanding is that they only look at if the values are correct or not.
    I know the question is a bit silly, but I am thinking if the time for doing the first table could be saved?
    Im not an examiner but my teacher who is very knowledgeable says they dont give a flying monkey about how you set it out AS LONG AS you set it out to show ,O, O2 , E & then your calculation for the sum of the totals, then show your reduction of the total.

    I put it into coulmn's with headings in the order aforementioned.
    Offline

    18
    ReputationRep:
    (Original post by Inges)
    Which paper did they use those words on may I ask?

    They usually use "Standard error" or "standard error of the mean" so seems unusual to me!
    they don't, they say something like in a sample the error is distributed such that...then they ay find prob that total error in 30 different things is whatever. you just go sum of 30 errors=nxmean so just bang in some CLT and ur done.
    Offline

    2
    ReputationRep:
    (Original post by Inges)
    I know right!
    Haha. Also I have a question regarding the 2015 question about the difference in the standard errors between the children and the staff, could you explain it to me? I'm a bit lost on that one. Thanks.
    Offline

    3
    ReputationRep:
    (Original post by physicsmaths)
    they don't, they say something like in a sample the error is distributed such that...then they ay find prob that total error in 30 different things is whatever. you just go sum of 30 errors=nxmean so just bang in some CLT and ur done.
    Again, which paper is this from? I have never seen a question of this sort?
    Offline

    2
    ReputationRep:
    (Original post by Inges)
    Im not an examiner but my teacher who is very knowledgeable says they dont give a flying monkey about how you set it out AS LONG AS you set it out to show ,O, O2 , E & then your calculation for the sum of the totals, then show your reduction of the total.

    I put it into coulmn's with headings in the order aforementioned.
    Okay, then I would just draw one table with columns O, E, and (O-E)^2/E and put the values of E directly into the table. So i guess no need to draw original table in the question with the expected frequency instead. Calculatiions of E needs to be shown?

    Thanks!
    Offline

    3
    ReputationRep:
    Has there ever been a goodness of fit test for the Normal Distribution? If so what paper
    Offline

    22
    ReputationRep:
    (Original post by econam)
    Has there ever been a goodness of fit test for the Normal Distribution? If so what paper
    Yes. June 2013.
    Offline

    3
    ReputationRep:
    (Original post by Geraer1)
    Okay, then I would just draw one table with columns O, E, and (O-E)^2/E and put the values of E directly into the table. So i guess no need to draw original table in the question with the expected frequency instead. Calculatiions of E needs to be shown?

    Thanks!
    No no, Sorry I think I misunderstood your question. You have to show the way you obtained the expected frequencies in a table like they have set out showing your Row total x Column total / grand total. then the expected frequency. After that, you can do what you said.

    This has been the case in a few papers, and some papers they just show the expected frequency without calculation. I haven't noticed a pattern in when they want you to show the working but id show it anyway to be safe!
    Offline

    22
    ReputationRep:
    (Original post by Geraer1)
    Okay, then I would just draw one table with columns O, E, and (O-E)^2/E and put the values of E directly into the table. So i guess no need to draw original table in the question with the expected frequency instead. Calculatiions of E needs to be shown?

    Thanks!
    That's fine. And no, you don't need to show calculation of E_i.
    Offline

    2
    ReputationRep:
    One thing to clarify: pmcc measures the linear association or linear correlation, because I just saw in a mark scheme which says linear association?
    Offline

    22
    ReputationRep:
    (Original post by paradoxequation)
    Haha. Also I have a question regarding the 2015 question about the difference in the standard errors between the children and the staff, could you explain it to me? I'm a bit lost on that one. Thanks.
    Children and adults have vastly different weights, so if you consider them in the same category, the variance from the mean is going to be quite large given how different their weights are. So the variance is larger, hence the standard deviation is larger and hence the standard error is larger since \text{s.e} = \frac{\sigma}{\sqrt{n}}. i.e: it is proportional to standard deviation.

    However, if you consider the children and adults in two separate categories, the variances from the mean are much smaller.
    Offline

    9
    ReputationRep:
    Is the spearman's rank wrong for June 2013 R Q3? They appear to have ordered the employees completely wrong!?

    Name:  JUNE 2013 R q3.jpg
Views: 108
Size:  353.5 KB
    Offline

    22
    ReputationRep:
    (Original post by Geraer1)
    One thing to clarify: pmcc measures the linear association or linear correlation, because I just saw in a mark scheme which says linear association?
    Definitely linear correlation. However, once you specify the word linear, correlation and association become the same thing. But in general, pmcc and spearman is always correlation and never association. Association is always for contingency tables.
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: June 29, 2016

University open days

  1. University of Bradford
    University-wide Postgraduate
    Wed, 25 Jul '18
  2. University of Buckingham
    Psychology Taster Tutorial Undergraduate
    Wed, 25 Jul '18
  3. Bournemouth University
    Clearing Campus Visit Undergraduate
    Wed, 1 Aug '18
Poll
How are you feeling in the run-up to Results Day 2018?

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.