+ Reply to Thread
Results 1 to 6 of 6

Thread: Basic question on Spearman's correlation coefficient

  1. #1
    Points: 842, Level: 15
    Level completed: 43%, Points required for next Level: 58

    Posts
    9
    Thanks
    3
    Thanked 1 Time in 1 Post

    Basic question on Spearman's correlation coefficient




    Hey guys,

    If I understand correctly, when calculating Spearman's correlation coefficient you just rank your variables, and then calculate Pearson's correlation coefficient for those new, ranked variables.

    However, while it is possible to calculate confidence intervals for Pearson, this is not possible for Spearman. Another procedure, canonical correlation analysis, also seems to be only possible to Pearson.

    My question is: why are these not possible for Spearman? What would go wrong when one just calculated the confidence intervals for Pearson, but using ranked data? Or do a canonical correlation analysis on ranked data?

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Basic question on Spearman's correlation coefficient

    I don't work with spearman's rho (there are a lot of spearman's and I assume you mean the ordinal measure of correlation) but it appears that it is possible to generate a CI for it using the Fischer transformation. See "Determing the Significance" below

    http://en.wikipedia.org/wiki/Spearma...on_coefficient

    This says the same thing

    http://www.statsdirect.com/help/defa...s/spearman.htm
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  3. The Following User Says Thank You to noetsi For This Useful Post:

    Hombre (02-24-2014)

  4. #3
    Points: 842, Level: 15
    Level completed: 43%, Points required for next Level: 58

    Posts
    9
    Thanks
    3
    Thanked 1 Time in 1 Post

    Re: Basic question on Spearman's correlation coefficient

    Ahhhh that's kind of embarrassing to get a wikipedia link (should've checked that better before asking :-)), but still very helpful, thanks!

  5. #4
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Basic question on Spearman's correlation coefficient

    I tend to be nervous about Wiki as it involves stats, because others have raised questions about accuracy at times. It is why I posted a second link.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  6. #5
    Devorador de queso
    Points: 95,922, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: Basic question on Spearman's correlation coefficient

    Quote Originally Posted by noetsi View Post
    I tend to be nervous about Wiki as it involves stats, because others have raised questions about accuracy at times. It is why I posted a second link.
    If it makes you feel better I'm actually typically not nervous about Wikipedia. Their math/stats material usually is pretty good! It's always better to look at another resource but I do use wikipedia as a resource more often than you might imagine.
    I don't have emotions and sometimes that makes me very sad.

  7. #6
    TS Contributor
    Points: 22,432, Level: 93
    Level completed: 9%, Points required for next Level: 918
    spunky's Avatar
    Location
    vancouver, canada
    Posts
    2,135
    Thanks
    166
    Thanked 537 Times in 431 Posts

    Re: Basic question on Spearman's correlation coefficient


    Quote Originally Posted by Dason View Post
    If it makes you feel better I'm actually typically not nervous about Wikipedia. Their math/stats material usually is pretty good! It's always better to look at another resource but I do use wikipedia as a resource more often than you might imagine.
    I have always assumed that Wikipedia is so awesome for math/stats stuff basically because: (a) it's not a very controversial area so you don't get people with weird ideas trying to sway your opinion one way or another. plus most people don't really study math at the level where you would want to make weird claims (b) because it deals with (mostly) well-known, verifiable truths. it doesn't take much to read and see quickly whether a certain theorem or claim is justified or not.

    all in all, Wikipedia = awesome for math/stats stuff.

    don't use it f
    for all your psychometric needs! https://psychometroscar.wordpress.com/about/

+ Reply to Thread

           




Posting Permissions

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






Advertise on Talk Stats