SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 


#Analytics
#SPSSStatistics
#Analyticstools
 View Only

  • 1.  Chi test correlation analysis and confounder effect elimination

    Posted 08/10/22 12:57 PM
    Hi, I am facing a problem...
    I have one independent ordinal variable (3 levels) and 1 categorical nominal dependant variable (yes/no). So it forms a 2x3 table. 
    When I run the correlation chi test square, I obviously don't receive a Fisher exact parameter and on the other hand, I can't use the Pearson Chi-square because of the violation of less than 20% of expected counts of less than 5.....
    What do you suggest I do in this situation? 
    and the last question, 
    how can I eliminate the potential confounder effect of two other independent variables (age and sex) on my correlation analysis? 

    Thank you so much for any help in this matter!
    Dan.

    ------------------------------
    Dan Feldman
    ------------------------------

    #SPSSStatistics


  • 2.  RE: Chi test correlation analysis and confounder effect elimination

    Posted 08/10/22 04:39 PM
    When I read your post, it occurred to me to suggest either the NOMREG or GENLOG procedure, because this is a modeling problem (i.e., the covariates you mentioned).  But I am by no means an expert in this area. I ran it by one of our statisticians. His reply:

    "In order to take additional predictors into account, he'd need to use a more general modeling framework such as a logistic regression model. In that framework, the ordinal predictor he's interested in would be handled in effectively the same way as a nominal three-level predictor. Statistics does not have any procedures that would provide exact p values; tests would be based on asymptomatic assumptions."

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



  • 3.  RE: Chi test correlation analysis and confounder effect elimination

    Posted 08/10/22 06:20 PM
    Thank you very much!!! 

    בתאריך יום ד׳, 10 באוג׳ 2022 ב-16:41 מאת Rick Marcantonio via IBM Community <Mail@connectedcommunity.org>:
    When I read your post, it occurred to me to suggest either the NOMREG or GENLOG procedure, because this is a modeling problem (i.e., the... -posted to the "SPSS Statistics" group
    IBM Community

    SPSS Statistics

    Post New Message
    Re: Chi test correlation analysis and confounder effect elimination
    Reply to Group Reply to Sender
    Rick Marcantonio
    Aug 10, 2022 4:39 PM
    Rick Marcantonio
    When I read your post, it occurred to me to suggest either the NOMREG or GENLOG procedure, because this is a modeling problem (i.e., the covariates you mentioned).  But I am by no means an expert in this area. I ran it by one of our statisticians. His reply:

    "In order to take additional predictors into account, he'd need to use a more general modeling framework such as a logistic regression model. In that framework, the ordinal predictor he's interested in would be handled in effectively the same way as a nominal three-level predictor. Statistics does not have any procedures that would provide exact p values; tests would be based on asymptomatic assumptions."

    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------
      Reply to Group Online   View Thread   Recommend   Forward  
    -------------------------------------------
    Original Message:
    Sent: Wed August 10, 2022 11:18 AM



     
    You are receiving this notification because you followed the 'Chi test correlation analysis and confounder effect elimination' 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


Related Content