LISTSERV mailing list manager LISTSERV 15.5

Help for SOCNET Archives


SOCNET Archives

SOCNET Archives


View:

Next Message | Previous Message
Next in Topic | Previous in Topic
Next by Same Author | Previous by Same Author
Chronologically | Most Recent First
Proportional Font | Monospaced Font

Options:

Join or Leave SOCNET
Reply | Post New Message
Search Archives


Subject: Re: Using node level regression analysis in Ucinet
From: Ian McCulloh <[log in to unmask]>
Reply-To:Ian McCulloh <[log in to unmask]>
Date:Fri, 9 Sep 2011 08:36:18 -0400
Content-Type:text/plain
Parts/Attachments:
Parts/Attachments

text/plain (64 lines)


*****  To join INSNA, visit http://www.insna.org  *****

I think the fundamental point is that OLS assumes independence, among other things. Residual analysis is really a tool to validate whether those assumptions hold.  We know already that independence is not a valid assumption.  The degree to which this may bias findings is unclear (to me at least).  I'd appreciate some discussion on this topic by the group.  Thanks.

Ian McCulloh

On Sep 9, 2011, at 8:29 AM, Philip Leifeld <[log in to unmask]> wrote:

> *****  To join INSNA, visit http://www.insna.org  *****
> 
> Dear Kamal,
> 
> How about simply transforming your node-level data in combination with your permutation approach, e.g. log(data) or by using a Box-Cox power transformation (if all else fails)? But be careful as this might change the interpretation of coefficients. Please correct me if I'm wrong.
> 
> Philip
> 
> 
> Am 07.09.2011 06:38, schrieb kamal badar:
>> *****  To join INSNA, visit http://www.insna.org  *****
>> 
>> Dear All,
>> 
>> According to Hanneman&  Riddle (2005), predicting and testing
>> hypotheses about actors non-relational attributes (e.g. their
>> performance) using a mix of relational (e.g. centrality) and
>> non-relational (e.g. gender) attributes, we might use node level
>> regression analysis in Ucinet (Tools>Testing
>> Hypotheses>Node-level>Regression). This process computes basic linear
>> multiple regression statistics by OLS, and estimate standard errors
>> and significance using the random permutations method for
>> constructing sampling distributions of R-squared and slope
>> coefficients. But how about the issue of shape of the distribution or
>> normality assumption? My data meets the first assumption of using
>> node level network regression but it reveals highly positively skewed
>> distribution when it is tested in SPSS. Can I still use node level
>> network regression? How can I defend it if some one asks about the
>> skewness of my data? Is there any other kind of regression in SPSS
>> which meets both assumptions?
>> 
>> Looking forward for insightful replies in this regard.
>> 
>> Regards
>> 
>> Kamal Badar Doctoral Student SOM.
>> 
>> _____________________________________________________________________
>> 
>> 
> 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.

_____________________________________________________________________
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.

Back to: Top of Message | Previous Page | Main SOCNET Page

Permalink



WWW.LISTS.UFL.EDU

CataList Email List Search Powered by the LISTSERV Email List Manager