+ Reply to Thread
Results 1 to 2 of 2

Thread: Non-parametric test (2 related samples)

  1. #1
    Points: 983, Level: 16
    Level completed: 83%, Points required for next Level: 17

    Thanked 0 Times in 0 Posts

    Non-parametric test (2 related samples)

    At present I am working on a hospital data set.

    I have categorised cases as either having Error or No Error.
    Another variable is the hospital name, of which there are eight names.

    I have run a chi-square test to observe if there is a significant association between error data and the hospital.

    However, I would also like to determine which hospital is responsible for errors.

    Is the Analyze > Non parametric tests > 2 related samples the appropriate test?

    The test does not seem to makes sense with variables mentioned, have I chosen the wrong test? Can anyone advise otherwise?

  2. #2
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    FC Schalke 04, Germany
    Thanked 639 Times in 601 Posts

    Re: Non-parametric test (2 related samples)

    I guess it's SPSS what you are using?
    You could have a look on the standardized adjusted
    residuals from your Chi² test. A value > |1,96| is
    an indication that a hospital deviates from expectation.

    Or, you could create 8 new grouping variables, each
    distinguishing between 1 hospital versus all other
    hospitals, and perform 8 crosstabs with chi².

    With kind regards


+ 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