+ Reply to Thread
Results 1 to 8 of 8

Thread: Low or high correlation?

  1. #1
    Points: 69, Level: 1
    Level completed: 38%, Points required for next Level: 31

    Posts
    6
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Low or high correlation?




    What does it mean when Pearson correlation gives a value of 0.7-0.8, while spearman correlation gives a value from 0.3-0.4? Pearson reports high correlation while Spearman report weak. Should I conclude that the correlation is weak using spearman since the data seems skewed?

  2. #2
    TS Contributor
    Points: 21,765, Level: 92
    Level completed: 42%, Points required for next Level: 585

    Location
    Israel
    Posts
    393
    Thanks
    0
    Thanked 7 Times in 7 Posts

    Re: Low or high correlation?

    Large differences between Spearman and Pearson can indicate the presence of outliers in your data

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

    John.Thomas (03-13-2014)

  4. #3
    R purist
    Points: 35,103, Level: 100
    Level completed: 0%, Points required for next Level: 0
    TheEcologist's Avatar
    Location
    United States
    Posts
    1,921
    Thanks
    303
    Thanked 607 Times in 341 Posts

    Re: Low or high correlation?

    Quote Originally Posted by WeeG View Post
    Large differences between Spearman and Pearson can indicate the presence of outliers in your data
    More general, it may indicate non-normality.
    The true ideals of great philosophies always seem to get lost somewhere along the road..

  5. The Following User Says Thank You to TheEcologist For This Useful Post:

    John.Thomas (03-13-2014)

  6. #4
    TS Contributor
    Points: 40,715, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Downloads
    gianmarco's Avatar
    Location
    Italy
    Posts
    1,369
    Thanks
    232
    Thanked 302 Times in 226 Posts

    Re: Low or high correlation?

    Hi!
    I believe that you should plot the data (i.e., scatterplot) to have a better idea of what is going on there.

    Gm
    http://cainarchaeology.weebly.com/

  7. The Following User Says Thank You to gianmarco For This Useful Post:

    John.Thomas (03-13-2014)

  8. #5
    Points: 69, Level: 1
    Level completed: 38%, Points required for next Level: 31

    Posts
    6
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Low or high correlation?

    After plotting the data, it showed outliners. Should spearman be used? Also should I consider using log?
    Last edited by John.Thomas; 03-13-2014 at 05:58 AM.

  9. #6
    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: Low or high correlation?

    There are various opinions about when you should and should not use these two. The one I was taught was that spearman was used for ordinal data and pearson for interval (pearson can not address non-linarity, spearman will work with non-linear data if it is monotonic in nature that is you have one rather than multiple directions for the curve).

    I have not heard that logs work to eliminate outliers - they stabilize variance. There are robust methods to address outliers, but they are not simple (such as M estimators).
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  10. #7
    TS Contributor
    Points: 40,715, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Downloads
    gianmarco's Avatar
    Location
    Italy
    Posts
    1,369
    Thanks
    232
    Thanked 302 Times in 226 Posts

    Re: Low or high correlation?

    @John.Thomas:
    Regardless the outlier(s), I would like to inspect the whole scatter of data points to have a broader idea of the correlation....
    http://cainarchaeology.weebly.com/

  11. #8
    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: Low or high correlation?


    One thing a scatterplot may catch is if there is non-linearity and if there is, is it monotonic or not. That will have a lot to do with what approach (pearson or spearman) you use.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

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

    John.Thomas (03-13-2014)

+ Reply to Thread

           




Tags for this 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