[R-sig-ME] Extracting higher level coefficients to replicate Douglas Luke's table

[Steven J. Pierce]

Richard,

Perhaps you can directly contact Douglas Luke to inquire about obtaining the
supporting materials. Here's a link to his webpage at his university:
http://publichealth.wustl.edu/people/scholardatabase/Pages/Luke.aspx 


Steven J. Pierce, Ph.D. 
Associate Director 
Center for Statistical Training & Consulting (CSTAT) 
Michigan State University 
E-mail: pierces1 at msu.edu 
Web: http://www.cstat.msu.edu 

-----Original Message-----
From: Richard Asturia [mailto:richard.asturia at gmail.com] 
Sent: Sunday, February 17, 2013 12:03 AM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] Extracting higher level coefficients to replicate
Douglas Luke's table

Dears, I have a very newby problem as I am migrating to lme4.

I am trying to replicate Douglas Luke's "Multilevel Modelling", table 2.9
(p.30-31) but using lmer and I don't know how to extract the coeficients of
the impact of variables in the 2nd level on betas.

Let me make it clearer by describing the situation. In the model number 3
in that table, the author models the following equations:

VotePct = b0 + b1 Party + b2 Money + r

                 b0 = y00 + y01 Acres + u0
                 b1 = y10 + y11 Acres + u1
                 b2 = y20 + y21 Acres + u2


The results are:

#Fixed Effects

For Intercept (b0)
Intercept (y00)           .1828  (.0205)
Acres (y01)                .0027  (.0005)

For Party slope (b1)
Party (y10)                 .5066   (.0215)
Acres (y11)                .-.0016   (.0004)

For Money slope (b2)
Money (y20)               .0049   (.0005)
Acres (y21)                 -.00002   (.0000)


But lmer only gives me the main coeficients, not for instance the specific
coeficient that for Acres on each beta.

Please, any tips on how to get that kind of results will be very welcome!

Richard

PS: btw, if someone can locate the support material of that book, it would
help a lot. The link provided in the book is broken.

	[[alternative HTML version deleted]]