# Read grouped beetle data from the web:
data="http://www.stat.ufl.edu/~aa/glm/data/Beetles2.dat"
beetle = read.table(data, header = T)

# Inspect grouped beetle data:
beetle

# Fit logistic regression with logdose as covariate:
fit=glm(cbind(dead,n-dead)~logdose,family=binomial,data=beetle)
summary(fit)


# Read ungrouped beetle data:
data="http://www.stat.ufl.edu/~aa/glm/data/Beetles.dat"
beetle.u = read.table(data, header = T)

# Inspect ungrouped beetle data:
head(beetle.u,n=10)
tail(beetle.u,n=10)

# Fit logistic regression model based on ungrouped data:
fit.u=glm(y~x,family=binomial,data=beetle.u)
#Info on how to use glm with different distributins/links:
#('Details' gives useful info for binomial data)
help("family")
summary(fit.u)
summary(fit)
# The estimates, standard errors, z values (Wald statistics) and P-values are the 
# same for grouped and ungrouped data. But all outputs are not the same.