Re: Interpreting Mediation results

Posted by peter link on Aug 18, 2014; 3:02pm
URL: http://spssx-discussion.165.s1.nabble.com/Interpreting-Mediation-results-tp5726809p5726992.html

Hi - Those are regression/path coefficients or products thereof and not effect sizes.

 

The Total Effect: The effect of X on Y. Often this is labeled as the c path.

The Direct Effect: The effect of X on Y adjusting for M. Often this is labeled as the c’ path.

The Indirect Effect: The product of the effects of X on M (a path) and M on Y (b path). Often this product is labeled as the ab path.

 

Hope this helps.

 

peter

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of psy_vm
Sent: Friday, August 15, 2014 11:56 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Interpreting Mediation results

 

I meant the sizes of the total, direct and indirect effects (.0465, .0059 and .0406 respectively). Are they Cohen's d or some other ones? I did mail him with this and the previous questions and he suggested I buy the book, which I just did. I read about the 6 different effects process gives at the bottom but I wanted to know about these other 3 ones. Makes sense? Vijaita Sent from Yahoo Mail on Android

 

---

From: peter link [via SPSSX Discussion] <[hidden email]>;
To: psy_vm <[hidden email]>;
Subject: Re: Interpreting Mediation results
Sent: Fri, Aug 15, 2014 11:26:51 PM

 

I’m not familiar with PROCESS but from the output it looks like an R-sq and Kappa-sq is provided. This is towards the bottom of the output. Is this different than what you are asking about?   By the way, Andrew Hayes has an email address listed on his website. Perhaps, go to his website and contact him directly.   Peter   From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of psy_vm Sent: Friday, August 15, 2014 3:48 PM To: [hidden email] Subject: Re: [SPSSX-L] Interpreting Mediation results   Hello, Please may you help me with an additional question? Do you know what effect size has Hayes used in his process? I've been through his book where he talks about the 6 different effect sizes that process gives in the result section: Partial, complete, r_square, kappa, etc. But which one has he used in the total, direct, indirect effects, etc. Would appreciate your help. Thanks, Vijaita Sent from Yahoo Mail on Android   From: peter link [via SPSSX Discussion] <[hidden email]>; To: psy_vm <[hidden email]>; Subject: Re: Interpreting Mediation results Sent: Thu, Jul 24, 2014 5:12:02 PM   The short answer is that if you are using a method such as bootstrapping to test the Indirect Effect, then you need to only find a significant ab-path. If you are using a method like The Test of Joint Significance (TJS), then you would look to see if both a and b are different from 0. In your example - The test of the indirect effect is under "Total, Direct, and Indirect Effects".  The CI for the Indirect Effect is [.0009,.1240] which doesn't include zero, so you would conclude that it is statistically significant (using bootstrapping criterion).  So, you could have an "a" or "b" path that is non-significant, but the "ab" path is significant. However, the Total Effect is non-significant, indicating that there isn't an effect to mediate, to begin with. Do you expect a Total Effect (should X predict Y)? The Direct and Indirect Effects have the same sign, so you aren't dealing with "Inconsistent Mediation" (a suppression effect).  So, you could talk about having an Indirect Effect but not Mediation.  Also, you should look at the size of the Indirect Effect.  It seems small - is this an important effect or is it trivial? pl -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of psy_vm Sent: Thursday, July 24, 2014 12:17 AM To: [hidden email] Subject: [SPSSX-L] Interpreting Mediation results Hello, I have a few questions regarding interpretation of my results for Mediation using Hayes' Process macro. The key point for finding significance is to see that the Class Interval does not include zero. However, in my mediation model, although the indirect effect CI doesn't include zero, the initial Models's CI (for bcope_ma, under Duration) includes zero. Should I still count my indirect effects as significant even though initial model for mediation is not significant? Results pasted below. Please Help! Thanks in advance! Model = 4     Y = qoltot     X = duration     M = bcope_ma Sample size          70 ************************************************************************** Outcome: bcope_ma Model Summary           R        R-sq          F            df1        df2          p       .1830      .0335     2.3550     1.0000    68.0000      .1295 Model                    coeff          se              t          p       LLCI ULCI constant    22.5169     1.3151    17.1223      .0000    19.8927    25.1411 duration     -.0388      .0253    -1.5346      .1295     -.0893      .0117 ************************************************************************** Outcome: qoltot Model Summary           R          R-sq          F          df1        df2          p       .3234      .1046     3.9140     2.0000    67.0000      .0247 Model                     coeff         se          t               p       LLCI ULCI constant      91.2951     9.5536     9.5560      .0000    72.2258   110.3643 bcope_ma    -1.0460      .3823    -2.7364      .0079    -1.8091     -.2830 duration      .0059          .0811      .0725      .9425     -.1559     .1677 ************************** TOTAL EFFECT MODEL **************************** Outcome: qoltot Model Summary           R        R-sq           F           df1        df2          p       .0674      .0045      .3102     1.0000    68.0000      .5794 Model                   coeff         se            t               p       LLCI ULCI constant    67.7413     4.3387    15.6134      .0000    59.0836    76.3990 duration      .0465      .0834      .5570      .5794     -.1200      .2129 ***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y      Effect         SE             t          p           LLCI       ULCI       .0465      .0834      .5570      .5794     -.1200      .2129 Direct effect of X on Y      Effect         SE          t             p         LLCI       ULCI       .0059      .0811      .0725      .9425     -.1559      .1677 Indirect effect of X on Y                     Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .0406      .0311      .0009      .1240 Partially standardized indirect effect of X on Y                    Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .0022      .0016      .0000      .0068 Completely standardized indirect effect of X on Y                    Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .0589      .0411      .0017      .1650 Ratio of indirect to total effect of X on Y                   Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .8736   149.0933      .1603    53.5001 Ratio of indirect to direct effect of X on Y                   Effect    Boot SE   BootLLCI   BootULCI bcope_ma     6.9093    17.9156     3.4423   275.8767 R-squared mediation effect size (R-sq_med)                  Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .0045      .0173     -.0184      .0543 Preacher and Kelley (2011) Kappa-squared              Effect    Boot SE   BootLLCI   BootULCI bcope_ma      .0601      .0401      .0066      .1692 Normal theory tests for indirect effect      Effect         se          Z          p       .0406      .0318     1.2753      .2022 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals:     1000 Level of confidence for all confidence intervals in output:    95.00 NOTE: Some cases were deleted due to missing data.  The number of such cases was:  33 -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Interpreting-Mediation-results -tp5726809.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD If you reply to this email, your message will be added to the discussion below: http://spssx-discussion.1045642.n5.nabble.com/Interpreting-Mediation-results-tp5726809p5726816.html To unsubscribe from Interpreting Mediation results, click here. NAML   View this message in context: Re: Interpreting Mediation results Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD If you reply to this email, your message will be added to the discussion below: http://spssx-discussion.1045642.n5.nabble.com/Interpreting-Mediation-results-tp5726809p5726985.html To unsubscribe from Interpreting Mediation results, click here. NAML

 

---

View this message in context: Re: Interpreting Mediation results
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD