+ Reply to Thread
Results 1 to 6 of 6

Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

  1. #1
    Forum Contributor
    Join Date
    11-11-2011
    Location
    London England
    MS-Off Ver
    Excel 2011 for Mac
    Posts
    157

    Question Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    Hi,

    I'm trying to run a Chi Squared test (for the first time and for the first time in Excel so there may be quite a few things that I have done wrong). I have followed the general formula through (I think, I found a lot of information on this site in terms of pointers: http://blog.excelmasterseries.com/20...d-easiest.html) and the histrogram distribution looks fairly normal, but when I follow everything through I have a fairly high Chi Squared Statistic in cell T30, resulting in a P Value of 0 (which is unhelpful to what I am trying to prove) in T32.

    The distribution in the Histogram data (around F22) looks fairly reasonable so I am struggling to figure quite where I have gone wrong. Do I need more bins, small bins, to balance my observations per bin up (so they are more balanced etc.). Any advice would be greatly appreciated.

    AFL_Chi_Squared.xls
    Last edited by mrvp; 05-17-2012 at 04:12 PM. Reason: Solved and very handy second opinion

  2. #2
    Valued Forum Contributor
    Join Date
    05-07-2012
    Location
    USA
    MS-Off Ver
    Excel 2007
    Posts
    354

    Re: Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    My notes/data begin in cell I50 of the "vertical" sheet. Hope this helps.

    Attachment 156192
    Regards,
    Vandan

  3. #3
    Valued Forum Contributor
    Join Date
    05-07-2012
    Location
    USA
    MS-Off Ver
    Excel 2007
    Posts
    354

    Re: Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    Excel 2003 format...

    Attachment 156376

  4. #4
    Forum Contributor
    Join Date
    11-11-2011
    Location
    London England
    MS-Off Ver
    Excel 2011 for Mac
    Posts
    157

    Re: Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    Seriously cool (and thanks for the spot on the older version of Excel) - I hadn't quite followed the explanation tutorial through properly (I didn't think I was far off) but that is a lot cleaner and more elegant way of doing things (and I can understand what I did wrong). How do you get the minitab bit up? Is that part of the statistics package in Excel? Does it make this test easier? Also what does this tell me exactly (I'm figuring the High Chi Square value shows a strong correlation between the data sets - in this case error of my predictions from a rating system against actual results of games based on a Standard Deviation of 33.239 and assuming a normal distribution)? - sorry if that last question was going out on a tangent a touch.

  5. #5
    Valued Forum Contributor
    Join Date
    05-07-2012
    Location
    USA
    MS-Off Ver
    Excel 2007
    Posts
    354

    Re: Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    1. Minitab is a stastical analysis program that I have. It is a widely used program. I used a snapshot from it to compare values excel produced.

    2. Chi Square test H0: Data follow a multinomial distribution with certain proportions
    Any P value below your alpha (usualy alpha = 0.05), we reject null hypothesis. In your case, since P-value is 0, we reject the null hypothesis

  6. #6
    Forum Contributor
    Join Date
    11-11-2011
    Location
    London England
    MS-Off Ver
    Excel 2011 for Mac
    Posts
    157

    Re: Chi Squared Test - Can anyone give me a second opinion on this (poor end values)

    Quote Originally Posted by vandan_tanna View Post
    1. Minitab is a stastical analysis program that I have. It is a widely used program. I used a snapshot from it to compare values excel produced.

    2. Chi Square test H0: Data follow a multinomial distribution with certain proportions
    Any P value below your alpha (usualy alpha = 0.05), we reject null hypothesis. In your case, since P-value is 0, we reject the null hypothesis
    Thanks I'd call this solved - and a handy maths lesson. On the basis that the null hypothesis is rejected I think this settles what I am trying to prove and that the data is approximately normally distributed (I'm going to go away and read about this and make my mind up).

+ Reply to Thread

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Search Engine Friendly URLs by vBSEO 3.6.0 RC 1