SPSS Statistics

 View Only

  • 1.  Mann Whitney test question

    Posted Tue February 08, 2022 01:57 PM
    Reply
    Hi everyone,

    just a quick question..

    I have a non parametric dataset and wanna apply Mann Whitney tests and Kruskal Wallis.
    My categorical variable in Mann whitney test is the Internet Addiction (N=9520) and not Internet addiction (N=4586), Total N=14206.

    When I am doing the Mann whitney test it seems that there are different number of participants in each comparison.

    for example: Stress with no Internet addiction (N=5469)
                         Stress with internet addiction (N=3298)
                                                   Total: 8767
                         Depression with no Internet addiction (N=6100)
                        Depression with internet addiction (N=3470)
                                                   Total: 9570

    In the same way it goes when I compare to other variables, e.g. depression,self efficacy etc., (the number of participants in each case is different) and when I apply the Kruskal Wallis test.

    Is this something normal, is this just missing items or something else went wrong?

    Your help is much appreciated!

    All the best

    #SPSSStatistics


  • 2.  RE: Mann Whitney test question

    Posted Tue February 08, 2022 02:08 PM
    Reply
    Sandy;

    Sounds like missing values to me.

    Note that for all NPAR TESTS, missing data have these options:

    • ANALYSIS. Exclude cases with missing values on a test-by-test basis. Cases with missing values for a variable that is used for a specific test are omitted from that test. On subcommands that specify several tests, each test is evaluated separately. This setting is the default.

    • LISTWISE. Exclude cases with missing values listwise. Cases with missing values for any variable that is named on any subcommand are excluded from all analyses.

    • INCLUDE. Include user-missing values. User-missing values are treated as valid values.


    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------

    Original Message


  • 3.  RE: Mann Whitney test question

    Posted Fri February 11, 2022 02:47 AM
    Reply
    Dear Rick Marcantonio,

    Thanks a lot for your help.

    Your advice was really helpful!



    Στις Τρί 8 Φεβ 2022 στις 9:10 μ.μ., ο/η Rick Marcantonio via IBM Community <Mail@connectedcommunity.org> έγραψε:
    Sandy; Sounds like missing values to me. Note that for all NPAR TESTS, missing data have these options: ANALYSIS. Exclude cases with... -posted to the "SPSS Statistics" group
    IBM Community

    SPSS Statistics

    Post New Message
    Re: Mann Whitney test question
    Reply to Group Reply to Sender
    Rick Marcantonio
    Feb 8, 2022 2:08 PM
    Rick Marcantonio
    Sandy;

    Sounds like missing values to me.

    Note that for all NPAR TESTS, missing data have these options:

    • ANALYSIS. Exclude cases with missing values on a test-by-test basis. Cases with missing values for a variable that is used for a specific test are omitted from that test. On subcommands that specify several tests, each test is evaluated separately. This setting is the default.

    • LISTWISE. Exclude cases with missing values listwise. Cases with missing values for any variable that is named on any subcommand are excluded from all analyses.

    • INCLUDE. Include user-missing values. User-missing values are treated as valid values.


    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------
      Reply to Group Online   View Thread   Recommend   Forward  

    Original Message



     
    You are receiving this notification because you followed the 'Mann Whitney test question' message thread. If you do not wish to follow this, please click here.

    Update your email preferences to choose the types of email you receive

    Unsubscribe from all participation emails



    Original Message:
    Sent: 2/8/2022 2:08:00 PM
    From: Rick Marcantonio
    Subject: RE: Mann Whitney test question

    Sandy;

    Sounds like missing values to me.

    Note that for all NPAR TESTS, missing data have these options:

    • ANALYSIS. Exclude cases with missing values on a test-by-test basis. Cases with missing values for a variable that is used for a specific test are omitted from that test. On subcommands that specify several tests, each test is evaluated separately. This setting is the default.

    • LISTWISE. Exclude cases with missing values listwise. Cases with missing values for any variable that is named on any subcommand are excluded from all analyses.

    • INCLUDE. Include user-missing values. User-missing values are treated as valid values.


    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------


Related Content