Re: Really Easy Question about Mediation (HELP!!)

Posted by on May 19, 2011; 6:10pm
URL: http://spssx-discussion.165.s1.nabble.com/Really-Easy-Question-about-Mediation-HELP-tp4405692p4410494.html

TBF et al.,

I wrote a quick & dirty primer on how to run mediation analyses in SPSS, but when I tried to send it to the listserv as an attachment, it got kicked back to me.  Here's the syntax that creates the coefficients necessary for most/all methods of mediation models.

 

In this case, the DV is DRUGUSEPOST.

The IV is GROUP.

The M is COPINGSKILLSw8.

The values of a, b, and c are taken from the standard regression equations, specifically the unstandardized beta weights and associated standard errors.

Here is example SPSS syntax for such a process:

Step I: 

 

REGRESSION

  /DESCRIPTIVES MEAN STDDEV CORR SIG N

  /MISSING LISTWISE

  /STATISTICS COEFF OUTS R ANOVA

  /CRITERIA=PIN(.05) POUT(.10)

  /NOORIGIN

  /DEPENDENT drugusepost

  /METHOD=ENTER group.

                   

Step II.

 

REGRESSION

  /DESCRIPTIVES MEAN STDDEV CORR SIG N

  /MISSING LISTWISE

  /STATISTICS COEFF OUTS R ANOVA

  /CRITERIA=PIN(.05) POUT(.10)

  /NOORIGIN

  /DEPENDENT  copingskillsw8

  /METHOD=ENTER group  .

 

Step III.

 

REGRESSION

  /DESCRIPTIVES MEAN STDDEV CORR SIG N

  /MISSING LISTWISE

  /STATISTICS COEFF OUTS R ANOVA

  /CRITERIA=PIN(.05) POUT(.10)

  /NOORIGIN

  /DEPENDENT drugusepost

Step I is used to obtain the value of c

  /METHOD=ENTER copingskillsw8 group.

 

And the statistical output:

 

Step I.

                                                                                Coefficients(a)

 

  

 

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta	B	Std. Error
1	(Constant)	32.895	2.762		11.909	.000
	Treatment Group	1.241	2.762	.066	.449	.655

Step II is used to obtain the value of a

a  Dependent Variable: drugusepost

 

Step II.

                                                                                Coefficients(a)

 

  

 

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta	B	Std. Error
1	(Constant)	4.226	.123		34.280	.000
	Treatment Group	.258	.123	.295	2.096	.042

a  Dependent Variable: copingskillsw8

Step III is used to obtain the value of b and c’

 

Step III.

                                                                                Coefficients(a)

 

  

 

 

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta	B	Std. Error
1	(Constant)	-8.920	12.909		-.691	.493
	Treatment Group	-1.316	2.623	-.070	-.502	.618
	copingskillsw8	9.894	2.996	.461	3.302	.002

a  Dependent Variable: drugusepost

 

HTH

 

-----Original Message-----
From: John F Hall <[hidden email]>
To: [hidden email]
Sent: Thu, May 19, 2011 1:49 pm
Subject: Re: Really Easy Question about Mediation (HELP!!)

There's a chapter on Causal Explanations in Marsh and Elliott, Exploring Data, followed by a bit on crosstabs with 3 variables (using SPSS 13).  I know the late Cathie Marsh discussed much of the logic of this with Jim Davis (see his Elementary Survey Analysis) and Joe Spaeth when they taught on SST=RC Summer Schools, and there is a fair bit in Rosenberg, The Logic of Survey Analysis.  Graham Kalton always swore by elaboration and there's a bit in Moser and Kalton as well (but no SPSS as it wasn't invented then).  We used the term "intervening variable" T in various relations with X and Y.

 

 

John F Hall

 

[hidden email]  

www.surveyresearch.weebly.com

 

 

 

 

 

 

-----Original Message-----
From: SPSSX(r) Discussion [[hidden email]] On Behalf Of R B
Sent: 19 May 2011 03:12
To: [hidden email]
Subject: Re: Really Easy Question about Mediation (HELP!!)

 

For those interested, this is a nice tutorial on testing for indirect

effects using the graphical interface in AMOS.

 

http://amosdevelopment.com/video/indirect/flash/indirect.html

 

Ryan

 

On Wed, May 18, 2011 at 2:12 AM, toph_bei_fong <[hidden email]> wrote:

> Hi everyone,

> 

> I am very, very new to mediation and could use some help. I know the steps

> to Baron and Kenny's method but am unsure of how to actually do them in

> SPSS.

> 

> I have two categorical IVs (which I have dummy coded), the interaction term

> of these categorical variables and three continuous mediators.

> 

> So, here are my questions:

> 

> 1) If I'm wanting to look at mediation of one of my IV effects, would I

> enter that variable in by itself in Step 1 as a predictor and leave the

> other IV out? Or can I assess both at the same time by entering both as my

> IVs? Or should I just enter the full model (IV1, IV2, IV1*IV2)?

> 

> 2) If I'm wanting to look at mediation at it pertains to my interaction

> term, do it enter the interaction term as the sole predictor in Step 1 or do

> I also enter IV1 and IV2 as covariates?

> 

> I just really need someone to walk me through these steps, but after that

> I'm golden.

> 

> Thank you!

> 

> tbf

> 

> --

> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Really-Easy-Question-about-Mediation-HELP-tp4405692p4405692.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