Overview
   • Regression Analysis: 
    –Mediation
    –Moderation
   • Non-parametric tests
    –When? Why? How?
          Example 1
  Bumble wants to know whether the 
  relationship between children’s negative 
  affect and self esteem can be explained by 
  children’s bad behavior. 
   Unfortunately, he had to take another vacation to 
   Hawaii, so he left before he could check his results. 
   It’s up to you to do it for him.  What do you need to 
   test for? 
        Mediation Model (Baron & Kenney)
                                  M
                 a                                 b
                                 c'
            X                                           Y
                                 c
                                       
                                                Mediation
        • Definition: 
               – The mediator (M) explains or accounts for the 
                   relationship between X & Y
        • How to test (packet 2, p.3)
               – Baron & Kenney Procedure
               – Sobel test (obsolete)
               – Bootstrapping 
        • Interpretation
               – Avoid causal language, instead say: 
                     •  “The pattern of correlations is consistent with mediation.” 
                     •  “The association between X & Y is mediated by M.
          Mediation
  •Is there evidence for mediation?
  •Path A: Negative affect  Bad behavior
    B =.024, p < .001  
  •Path B: Bad behavior  Self Esteem
    B = -.649, p = .024
  •Path C: Negative affect  Self Esteem
    C: B = -.111, p < .001
    C' : B = -.096, p < .001