Displaying 1 result from an estimated 1 matches for "factor_glm_2".
2008 Nov 16
1
confint.glm(...) fails for binomial count data format
...le.factor)
## sum(womensrole$agree)
## sum(womensrole$disagree)
##Two dataset will report same prediction, Chisq and different sample
size, residual deviance, ...
fm2 <- cbind(agree,disagree) ~ sex * education;
womensrole_glm_2 <- glm(fm2, data = womensrole, family = binomial());
womensrole.factor_glm_2 <- glm(agree~sex*education, data =
womensrole.factor, family = binomial());
## Same prediction
myplot <- function(role.fitted) {
f <- womensrole$sex == "Female"
plot(womensrole$education, role.fitted, type = "n",
ylab = "Probability of agreeing",
xlab = &q...