significance of mediation effect - Sobel Z

[MKA-4]

Hi listers;
I have a general statistical question. I wonder if it is correct to make comments on the mediation effect in a regression equation, without making Sobel z-test for its significance. I have recently seen in an article such a situation.  Like that: A predicts B; when C is added to the equation, the effect of A on B weakens and somebody comments that C is partial mediatior. If the effect of A on B becomes insignificant, it is commented that C is full mediator without making significance test of Sobel Z.
My second question is; I wonder if there is a way in SPSS to make Sobel Z test or an alternative method to test the significance of mediation effect. Becuse it is a bit time consuming to calculate Sobel Z in hand.

Thanks in advance
Mehmet Karakus

====================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

[Sara House]

Mehmet -

I'm not sure about a way to make SPSS calculate Sobel's Z, but Kris Preacher has a website that will calculate Sobel's Z for you:

http://people.ku.edu/~preacher/sobel/sobel.htm

He also has some power analysis calculators and a few other goodies.  It's a fun website to check out.

Sara

Sara M. House, M.A.
Adjunct Faculty
Loyola University Chicago, Psychology Department
Email: [hidden email]
Teaching: Research Methods, Psychology & Law

=====================
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

[Jeff Stuewig]

        As Sara said Preacher's web site is a very good resource.  He also
has macros (for SAS and SPSS) that will use bootstrapping to compute
confidence intervals which seems to be a better test that just the pure
sobel test.

        Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
strategies for assessing and comparing indirect effects in multiple mediator
models. Behavior Research Methods, 40, 879-891.

http://www.people.ku.edu/~preacher/
http://www.comm.ohio-state.edu/ahayes/SPSS%20programs/indirect.htm




> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
MKA
> Sent: Wednesday, October 01, 2008 7:47 AM
> To: [hidden email]
> Subject: significance of mediation effect - Sobel Z
>
> Hi listers;
> I have a general statistical question. I wonder if it is correct to make
> comments on the mediation effect in a regression equation, without making
> Sobel z-test for its significance. I have recently seen in an article such
a
> situation.  Like that: A predicts B; when C is added to the equation, the
> effect of A on B weakens and somebody comments that C is partial
mediatior.
> If the effect of A on B becomes insignificant, it is commented that C is
> full mediator without making significance test of Sobel Z.
> My second question is; I wonder if there is a way in SPSS to make Sobel Z
> test or an alternative method to test the significance of mediation
effect.

> Becuse it is a bit time consuming to calculate Sobel Z in hand.
>
> Thanks in advance
> Mehmet Karakus
>
> =======
> 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

[Swank, Paul R]

Absolutely. Sobel's test lacks power and depends on normality
assumptions. Bootstrapping has more power and is not dependent on
normality assumptions.

Paul R. Swank, Ph.D
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center
Houston, TX 77038


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Jeff Stuewig
Sent: Wednesday, October 01, 2008 9:19 AM
To: [hidden email]
Subject: Re: significance of mediation effect - Sobel Z


        As Sara said Preacher's web site is a very good resource.  He
also
has macros (for SAS and SPSS) that will use bootstrapping to compute
confidence intervals which seems to be a better test that just the pure
sobel test.

        Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and
resampling
strategies for assessing and comparing indirect effects in multiple
mediator
models. Behavior Research Methods, 40, 879-891.

http://www.people.ku.edu/~preacher/
http://www.comm.ohio-state.edu/ahayes/SPSS%20programs/indirect.htm




> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf
Of
MKA
> Sent: Wednesday, October 01, 2008 7:47 AM
> To: [hidden email]
> Subject: significance of mediation effect - Sobel Z
>
> Hi listers;
> I have a general statistical question. I wonder if it is correct to
make
> comments on the mediation effect in a regression equation, without
making
> Sobel z-test for its significance. I have recently seen in an article
such
a
> situation.  Like that: A predicts B; when C is added to the equation,
the
> effect of A on B weakens and somebody comments that C is partial
mediatior.
> If the effect of A on B becomes insignificant, it is commented that C
is
> full mediator without making significance test of Sobel Z.
> My second question is; I wonder if there is a way in SPSS to make
Sobel Z
> test or an alternative method to test the significance of mediation
effect.
> Becuse it is a bit time consuming to calculate Sobel Z in hand.
>
> Thanks in advance
> Mehmet Karakus
>
> =======
> 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

=====================
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