cautres {gnm} | R Documentation |
A 4-way contingency table of vote by class by religion in four French elections
data(cautres)
A table of counts, with classifying factors vote
(levels
1:2
), class
(levels 1:6
) and religion
(levels 1:4
) and election
(levels 1:4
).
Bruno Cautres
Cautres, B, Heath, A F and Firth, D (1998). Class, religion and vote in Britain and France. La Lettre de la Maison Francaise 8.
set.seed(1) data(cautres) ## Fit a "double UNIDIFF" model with the religion-vote and class-vote ## interactions both modulated by nonnegative election-specific multipliers doubleUnidiff <- gnm(Freq ~ election*vote + election*class*religion + Mult(Exp(-1 + election), religion:vote) + Mult(Exp(-1 + election), class:vote), family = poisson, data = cautres) ## Deviance should be 133.04 ## Examine the multipliers of the class-vote log odds ratios ofInterest(doubleUnidiff) <- pickCoef(doubleUnidiff, "Mult2.*election") coef(doubleUnidiff) ## Mult2.Factor1.election1 Mult2.Factor1.election2 ## -0.5724370 0.1092972 ## Mult2.Factor1.election3 Mult2.Factor1.election4 ## -0.1230682 -0.2105843 ## Re-parameterize by setting Mult2.Factor1.election1 to zero getContrasts(doubleUnidiff, ofInterest(doubleUnidiff)) ## estimate SE quasiSE quasiVar ## Mult2.Factor1.election1 0.0000000 0.0000000 0.22854380 0.052232270 ## Mult2.Factor1.election2 0.6817370 0.2401642 0.07395880 0.005469905 ## Mult2.Factor1.election3 0.4493740 0.2473519 0.09475932 0.008979329 ## Mult2.Factor1.election4 0.3618262 0.2534753 0.10934823 0.011957035 ## Same thing but with election 4 as reference category: getContrasts(doubleUnidiff, rev(ofInterest(doubleUnidiff))) ## estimate SE quasiSE quasiVar ## Mult2.Factor1.election4 0.00000000 0.0000000 0.10934823 0.011957035 ## Mult2.Factor1.election3 0.08754785 0.1446834 0.09475932 0.008979329 ## Mult2.Factor1.election2 0.31991082 0.1320023 0.07395880 0.005469905 ## Mult2.Factor1.election1 -0.36182617 0.2534753 0.22854380 0.052232270