| yaish {gnm} | R Documentation |
A 3-way contingency table of father/son pairs, classified by father's
social class (orig), son's social class (dest) and son's
education level (educ).
data(yaish)
A table of counts, with classifying factors educ (levels 1:5),
orig (levels 1:7) and dest (levels 1:7).
David Firth
Yaish, M (1998). Opportunities, Little Change. Class Mobility in Israeli Society: 1974-1991. D.Phil. Thesis, Nuffield College, University of Oxford.
set.seed(1)
data(yaish)
## Fit the "UNIDIFF" mobility model across education levels
unidiff <- gnm(Freq ~ educ:orig + educ:dest +
Mult(Exp(-1 + educ), orig:dest), family = poisson,
data = yaish)
## Deviance should be 208.3
## Look at the multipliers of the orig:dest association:
coefs.of.interest <- grep("Mult1.*educ", names(coef(unidiff)))
coef(unidiff)[coefs.of.interest]
## Mult1.Factor1.educ1 Mult1.Factor1.educ2 Mult1.Factor1.educ3
## 1.9853067 1.7685557 1.2540751
## Mult1.Factor1.educ4 Mult1.Factor1.educ5
## 0.9551895 -0.2525413
## Get standard errors for contrasts with educ5
getContrasts(unidiff, coefs.of.interest)
## estimate se
## Mult1.Factor1.educ1 2.237848 0.9411054
## Mult1.Factor1.educ2 2.021097 0.9434948
## Mult1.Factor1.educ3 1.506616 0.9535574
## Mult1.Factor1.educ4 1.207731 0.9780784
## Mult1.Factor1.educ5 0.000000 0.0000000
## Table of model residuals
residuals(unidiff)