+ Reply to Thread
Results 1 to 5 of 5

Thread: Comparing 2 percentages - how?

  1. #1
    Points: 2,705, Level: 31
    Level completed: 70%, Points required for next Level: 45

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Comparing 2 percentages - how?




    Hi again guys,
    I need to compare two percentages.

    I have the percentage 12% and the percentage 38.5%

    I need to state if they are significantly different.

    I've been told I need to do a chi-square test but I'm not sure how exactly I'm meant to do that.

    Any advice would be much appreciated! Many thanks in advance

    Laura

  2. #2
    R purist
    Points: 18,370, Level: 86
    Level completed: 4%, Points required for next Level: 480
    TheEcologist's Avatar
    Location
    The Netherlands.
    Posts
    1,590
    Thanks
    201
    Thanked 409 Times in 227 Posts
    Quote Originally Posted by laura marie View Post
    Hi again guys,
    I need to compare two percentages.

    I have the percentage 12% and the percentage 38.5%

    I need to state if they are significantly different.

    I've been told I need to do a chi-square test but I'm not sure how exactly I'm meant to do that.

    Any advice would be much appreciated! Many thanks in advance

    Laura
    You can use an exact Binomial test;

    Online reference:

    http://en.wikipedia.org/wiki/Binomial_test.

    SPSS;
    http://academic.udayton.edu/gregelve...ominaldata.htm

    Other references:

    Clopper, C. J. & Pearson, E. S. (1934). The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika, 26, 404413.

    William J. Conover (1971), Practical nonparametric statistics. New York: John Wiley & Sons. Pages 97104.

    Myles Hollander & Douglas A. Wolfe (1973), Nonparametric statistical inference. New York: John Wiley & Sons. Pages 1522.



    good luck,
    The true ideals of great philosophies always seem to get lost somewhere along the road..

  3. #3
    Points: 2,705, Level: 31
    Level completed: 70%, Points required for next Level: 45

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Thank you for your reply, unfortunately I have read and re-read the links thoroughly and still don't know how I am meant to compute this.

    I only have these four numbers.

    The 38.5% is out of 52 participants.
    The 12% is out of 114 participants.

    I am using SPSS but I don't know how to arrange the numbers to run the test. It is very very frustrating!

  4. #4
    R purist
    Points: 18,370, Level: 86
    Level completed: 4%, Points required for next Level: 480
    TheEcologist's Avatar
    Location
    The Netherlands.
    Posts
    1,590
    Thanks
    201
    Thanked 409 Times in 227 Posts
    Quote Originally Posted by laura marie View Post
    Thank you for your reply, unfortunately I have read and re-read the links thoroughly and still don't know how I am meant to compute this.

    I only have these four numbers.

    The 38.5% is out of 52 participants.
    The 12% is out of 114 participants.

    I am using SPSS but I don't know how to arrange the numbers to run the test. It is very very frustrating!
    Lets see if this helps:

    You're sample size is probably sufficient (this is subjective) to allow the use of a Chi-squared and/or G-test.

    These test might be less difficult to compute and are explained in great detail on the web.

    Heres a resource:
    http://udel.edu/~mcdonald/statgtestind.html
    The true ideals of great philosophies always seem to get lost somewhere along the road..

  5. #5
    TS Contributor
    Points: 8,309, Level: 61
    Level completed: 53%, Points required for next Level: 141

    Posts
    858
    Thanks
    0
    Thanked 85 Times in 84 Posts

    Quote Originally Posted by laura marie View Post
    Thank you for your reply, unfortunately I have read and re-read the links thoroughly and still don't know how I am meant to compute this.

    I only have these four numbers.

    The 38.5% is out of 52 participants.
    The 12% is out of 114 participants.

    I am using SPSS but I don't know how to arrange the numbers to run the test. It is very very frustrating!
    You can do this in Excel (sorry I don't know SPSS). A very simple-minded Excel worksheet is attached, pardon its look.

    You know that 38.5% of 52 participants = 20 are "successes" in row 1
    You know that 12% of 114 participants = 14 are "successes" in row 2

    (See the ACTUAL table, rows 1-3)

    Thus you know that 34 of 166 participants (20.5%) are successes overall. So using this overall % of successes, you can determine the EXPECTED table (see rows 5-7)
    eg 20.5% of 52 participants = expected 10.6506 "successes"
    and 20.5% of 114 participants = expected 23.3494 "successes"

    Chi-square test compares the actual successes vs expected successes in row 1, and in row 2 (alternatively, you can say it compares the actual non-successes vs expected non-successes, since successes + non-successes = participants).

    Excel has a CHITEST function (see cell A12).
    Attached Files

+ Reply to Thread

           




Similar Threads

  1. Comparing percentages
    By amperes in forum Statistics
    Replies: 10
    Last Post: 10-11-2011, 03:31 PM
  2. comparing percentages
    By agabo in forum Statistics
    Replies: 5
    Last Post: 05-28-2011, 08:03 PM
  3. Comparing percentages
    By Em_J in forum Statistics
    Replies: 3
    Last Post: 11-27-2010, 08:32 AM
  4. Comparing percentages
    By lauri195 in forum Statistics
    Replies: 7
    Last Post: 07-19-2010, 07:48 PM
  5. Comparing two percentages for significance
    By breiteyes55 in forum Statistics
    Replies: 1
    Last Post: 04-16-2007, 09:48 AM

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