[R] ergm
Iasonas Lamprianou
lamprianou at yahoo.com
Thu Sep 23 08:51:43 CEST 2010
Dear colleagues,
I have another question, which, I think cannot be answered easily by the manual. What is the effect of including both nodefactor("Gender") and nodematch("Gender",diff=TRUE)) for the same variable in the model? Judging from the output (please see below), you cant have estimates for both for boys and girls forthe nodematch command, but I thought that the nodematch measures for boys and girls were not related. Does the nodematch command show the relative frequency of matching boy-boy against girl-girl, so one of the two categories has to be a reference category? Thank you
Formula: ng ~ edges + nodefactor("Gender") + nodematch("LEA", diff = TRUE) +
nodematch("Gender", diff = TRUE)
Newton-Raphson iterations: 5
Maximum Likelihood Results:
Estimate Std. Error MCMC s.e. p-value
edges -2.098842 0.033393 NA <1e-04 ***
nodefactor.Gender.Girl 0.612314 0.021956 NA <1e-04 ***
nodematch.LEA.1 0.018116 0.032228 NA 0.5740
nodematch.LEA.3 -0.002688 0.154005 NA 0.9861
nodematch.LEA.4 0.103054 0.043836 NA 0.0187 *
nodematch.LEA.5 0.016844 0.033565 NA 0.6158
nodematch.LEA.6 -0.086837 0.100466 NA 0.3874
nodematch.Gender.Boy 0.593721 0.038632 NA <1e-04 ***
nodematch.Gender.Girl NA 0.000000 NA NA
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
For this model, the pseudolikelihood is the same as the likelihood.
Null Deviance: 92092 on 66430 degrees of freedom
Residual Deviance: 67948 on 66421 degrees of freedom
Deviance: 24144 on 9 degrees of freedom
AIC: 67966 BIC: 68048
Dr. Iasonas Lamprianou
Assistant Professor (Educational Research and Evaluation)
Department of Education Sciences
European University-Cyprus
P.O. Box 22006
1516 Nicosia
Cyprus
Tel.: +357-22-713178
Fax: +357-22-590539
Honorary Research Fellow
Department of Education
The University of Manchester
Oxford Road, Manchester M13 9PL, UK
Tel. 0044 161 275 3485
iasonas.lamprianou at manchester.ac.uk
More information about the R-help
mailing list