***** To join INSNA, visit http://www.insna.org *****
Well, as a veteran UCINET user and huge fan, I will say I think the
easiest thing to do here is to use the R language. Network programs,
which will do exactly what you want given your description below, are
available in the STATNET and SNA packages in R. This is what I did when
I needed to capture point values under lots of simulated results in
order to estimate a pseudo-standard error. It is a trivial exercise in
R, once you get over the learning curve of mastering the language itself.
And R is free (open source, even).
-David
Carolyn B wrote:
> ***** To join INSNA, visit http://www.insna.org ***** My only worry
> doing it this way (mind you that was the solution I finally used as I
> couldn't figure out another way to test interactions) is violating the
> independence assumption when using regression, etc. in spss/sas rather
> than how network programs do it (for regression that is, presuming I
> understand it correctly).
>
> My question to others - is this an issue we need to take into
> consideration???
>
> Thanks,
> Carolyn
>
> ------------------------------------------------------------------------
> Date: Sun, 5 Jul 2009 01:13:38 -0400
> From: [log in to unmask]
> Subject: Re: [SOCNET] mediated effects using Sobel
> To: [log in to unmask]
>
> ***** To join INSNA, visit http://www.insna.org ***** Manju,
>
> To my knowledge, UCINET does not have the capability to produce the
> standard errors needed in the Sobel test. My suggestion would be to
> calculate your variables of interest in UCINET, then copy/export each
> of these into SPSS to run the series of regressions. For example, if
> you expect that betweeness centrality functions as a mediator,
> calculate this variable for your respondents in UCINET, copy the whole
> column for the variable into your SPSS file with your other variables,
> then run your regressions.
>
> Hope this helps,
>
> Jessica
>
>
> Jessica R. Methot
> PhD Student - Department of Management
> Warrington College of Business
> University of Florida
> Gainesville, FL 32611-7165
> Office: (352) 392-0102
> [log in to unmask]
>
>
> -----Original Message-----
> From: Social Networks Discussion Forum on behalf of Manju Ahuja
> Sent: Sat 7/4/2009 11:04 PM
> To: [log in to unmask]
> Subject: [SOCNET] mediated effects using Sobel
>
> ***** To join INSNA, visit http://www.insna.org *****
>
> Does anyone know of a way to perform a Sobel (1982) as a test of
> mediation while using UCINET? We have the formula to manually compute
> this statistic but it requires the standard errors of each beta
> coefficient. This is of course reported in SPSS but we cannot find a
> way to get this information in UCINET. If you know of a way to get
> this information, we would appreciate hearing about it.
>
> Thanks!
>
> Manju Ahuja
>
> Manju K. Ahuja, Ph.D.
> Professor and Chair
> Computer Information Systems
> College of Business
> University of Louisville
> Louisville, KY 40292
> Phone: (502) 852-4678
> Fax: (502) 852-4799
> [log in to unmask]
> http://cbpa.louisville.edu/profile/Ahuja.html
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
>
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send an
> email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
> ------------------------------------------------------------------------
> Insert movie times and more without leaving HotmailŪ. See how.
> <http://windowslive.com/Tutorial/Hotmail/QuickAdd?ocid=TXT_TAGLM_WL_HM_Tutorial_QuickAdd_062009>_____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send an
> email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
--
--------------------
David Krackhardt, Professor of Organizations, Executive Editor of JoSS
Heinz College of Public Policy and Management, and
The Tepper School of Business
Carnegie Mellon University
Pittsburgh, PA 15213
412-268-4758
website: www.andrew.cmu.edu/~krack
(Erdos#=2)
_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.
|
