+ Reply to Thread
Results 1 to 4 of 4

Thread: Correlation question

  1. #1
    Points: 530, Level: 10
    Level completed: 60%, Points required for next Level: 20

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Correlation question



    This might be a stupid question but can I test if A has correlation with x, y or z when x, y and z are percentages? For example:

    Code: 
    A	x	y	z
    0,1	0,8	0,1	0,1
    0,03	0,7	0,05	0,25
    0,05	0,6	0,05	0,35
    0,09	0,6	0,01	0,39
    0,04	0,7	0,04	0,26
    0,02	0,95	0,05	0
    0,06	0,65	0,3	0,05

    My problem is that I dont know is it a problem that when you for example increase x then both y or z or one of them need to decrease.

  2. #2
    IBM Rules
    Points: 12,754, Level: 73
    Level completed: 76%, Points required for next Level: 96

    Posts
    2,485
    Thanks
    108
    Thanked 365 Times in 353 Posts

    Re: Correlation question

    You can test interval data with correlations and percentages commonly are that. But I am confused why you apear to have two values in each cell. Correlation is only between single values in a cell (or all the types I know do anyhow).

  3. #3
    TS Contributor
    Points: 6,533, Level: 52
    Level completed: 92%, Points required for next Level: 17
    Lazar's Avatar
    Location
    Sydney
    Posts
    660
    Thanks
    109
    Thanked 164 Times in 149 Posts

    Re: Correlation question

    Hullu could be German where a commar is used to denote decimal places rather than a dot.

  4. #4
    Bhoot
    Points: 1,275, Level: 19
    Level completed: 75%, Points required for next Level: 25

    Posts
    1,758
    Thanks
    40
    Thanked 124 Times in 106 Posts

    Re: Correlation question


    Quote Originally Posted by hullu1234 View Post
    This might be a stupid question but can I test if A has correlation with x, y or z when x, y and z are percentages? For example:

    Code: 
    A    x    y    z
    0,1    0,8    0,1    0,1
    0,03    0,7    0,05    0,25
    0,05    0,6    0,05    0,35
    0,09    0,6    0,01    0,39
    0,04    0,7    0,04    0,26
    0,02    0,95    0,05    0
    0,06    0,65    0,3    0,05
    My problem is that I dont know is it a problem that when you for example increase x then both y or z or one of them need to decrease.
    What I can see from your data is X+Y+Z=1(100 %). So, it not a problem that when you for example increase x then both y or z or one of them need to decrease.
    For statistical analysis you just need to consider two variable.( eg: X &Y)
    In the long run, we're all dead.

+ 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