SPSS Statistics

SPSS Statistics

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

 View Only

  • 1.  Logistic regression - Confounders

    Posted Fri May 15, 2020 11:57 PM
    Hi, does anyone know how to control for counfounders (e.g. age, sex and race) in a binary logistic regression model?
    Also, when we adjust for the confounders, does this mean that they are not included in the model to predict an outcome? Or does it mean that include the confounders in the model, even though they are not always significant, because they are hypothetically supposed to be predictors and so they cannot just be removed from the model?
    Many thanks.

    ------------------------------
    Michelle Tan
    ------------------------------

    #SPSSStatistics


  • 2.  RE: Logistic regression - Confounders

    Posted Fri May 29, 2020 09:55 AM
    I would treat this the same as for ordinary regression.  I would include the confounders in the equation, but I would examine how the interesting coefficients change with and without the confounders.  You might want to do the predicted calculations holding the confounders constant.  If you use GENLIN, you can calculate estimated marginal means.

    ------------------------------
    Jon Peck
    ------------------------------

    Original Message