[R] SEM - standardized path coefficients? - significant style

John Fox jfox at mcmaster.ca
Thu Sep 20 17:02:39 CEST 2007


Dear Steve,

I'll file this suggestion for when I next revise the sem package. I'm
unlikely to get to it soon since I rarely use the graph-drawing program dot
and would have to figure out how to do what you want. As well, I think that
I'd more likely allow the user to specify the line type for each edge in the
graph rather than hard-coding this to significance/non-significance.

Regards,
 John

--------------------------------
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox 
-------------------------------- 

> -----Original Message-----
> From: Steve Powell [mailto:steve at promente.org] 
> Sent: Thursday, September 20, 2007 3:40 AM
> To: 'John Fox'
> Cc: r-help at stat.math.ethz.ch; eso at promente.org
> Subject: RE: [R] SEM - standardized path coefficients? - 
> significant style
> 
> Dear John
> Being an R user is like having Christmas every day and having 
> a direct line to Santa for present suggestions. 
> How about if path.diagram.sem had the ability to print 
> non-significant paths in, for example dotted style? I had a 
> go myself but couldn't see how to get 
> (summary(model)$coeff)$"Pr(>|z|)" into the loop.
> Very best wishes
> Steve Powell
> 
>  
> proMENTE social research
> research | evaluation | training & consulting Kranjčevićeva 
> 35, 71000 Sarajevo
> mobile: +387 61 215 997 | office: +387 33 556 865 | fax: +387 
> 33 556 866
> skype: stevepowell99
> www.promente.org  |  www.mojakarijera.com  |  www.psih.org  
> 



More information about the R-help mailing list