Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: SimpleSlopes in Multilevel Models


From   Cameron McIntosh <[email protected]>
To   STATA LIST <[email protected]>
Subject   RE: st: SimpleSlopes in Multilevel Models
Date   Sun, 5 Feb 2012 16:35:17 -0500

If -zpos- is metric, you can generally do:
b[ZSchaufelishort] | zpos value = b[ZSchaufelishort] + b[EExPOS]*zpos value

for as many different values as you want. I assume that can be implemented via -lincom- and would invoke the ACM.

Cam

> Date: Sun, 5 Feb 2012 15:55:33 -0500
> Subject: Re: st: SimpleSlopes in Multilevel Models
> From: [email protected]
> To: [email protected]
> 
> If -zpos- is a binary variable with values 0/1, then the relation for
> low levels is given by the coefficient itself, and for high levels, by
> 
> lincom _b[ ZSchaufelishort ] + _b[ EExPOS ]
> 
> which would invoke the covariance matrix behind the scenes. This can
> also be done with -margins-, and the corresponding plots, with
> -marginsplot-.
> 
> On Sun, Feb 5, 2012 at 2:03 PM, Kerstin Alfes <[email protected]> wrote:
> > Thanks for the references, Cam. I will look into them!
> >
> > Stan, I would like to graph the interaction and also calculate whether the relationship between ZSchaufelishort and logperformance is significant for high and/or low levels of ZPOS. If there is a way to do this without the ACM, that would help a lot.
> >
> > Best wishes
> > Kerstin
> >
> >
> > -------- Original-Nachricht --------
> >> Datum: Sun, 5 Feb 2012 10:21:14 -0500
> >> Von: Stas Kolenikov <[email protected]>
> >> An: [email protected]
> >> Betreff: Re: st: SimpleSlopes in Multilevel Models
> >
> >> On Sun, Feb 5, 2012 at 9:31 AM, Cameron McIntosh <[email protected]>
> >> wrote:
> >> > If you want to dissect the interaction effects further, yes you would
> >> need to estimate and test simple slopes, for which you would need the
> >> covariance between the main effect and interaction coefficients as you say. I
> >> don't know how to request the ACM in Stata, but tech support should be able to
> >> help you.
> >>
> >> I am pretty sure one won't need to try to get the covariance matrices
> >> (which are available as -e(V)-), as there is little you cannot do
> >> between -test-, -testnl-, -lincom- and -nlcom-. Matrix code will be
> >> ugly, and you need to master the convenience tools, anyway. What is
> >> involved in "estimating and testing simple slopes"?
> >>
> >> --
> >> Stas Kolenikov, also found at http://stas.kolenikov.name
> >> Small print: I use this email account for mailing lists only.
> >>
> >> *
> >> *   For searches and help try:
> >> *   http://www.stata.com/help.cgi?search
> >> *   http://www.stata.com/support/statalist/faq
> >> *   http://www.ats.ucla.edu/stat/stata/
> >
> > --
> > Kerstin Alfes
> >
> > Kettelerweg 24
> > 57482 Wenden
> > Germany
> > Phone: +49-(0)2762-2766
> > Mobile: +49-(0)174-160 15 44
> >
> > Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
> > belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> 
> 
> 
> -- 
> Stas Kolenikov, also found at http://stas.kolenikov.name
> Small print: I use this email account for mailing lists only.
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index