I would like to hear of possible solutions to the following problem. I want to integrate the confintcommand with the general logistic regression glm command on R so that it can be default provide the confidence intervals when runing logistic regression models. I am posting another discussion post that refers to confint and it is a useful link to avoid any misunderstandings as I rarely post questions here.
Computing Confidence Intervals for Coefficients in Logistic Regression
I tried the following code for generating the effects plot but it did not work due to the lack of confidence intervals:
M1<-glm(formula=approval_binary~compliance+militarism2+internationalism2+political_views+culture+income+education+language+religion+gender+age,family="binomial", data=dataset1)
summary(M1)
effects_M1=margins(M1)
print(effects_M1)
plot(effects_M1)
effects_M1=summary(effects_M1)
ggplot(data=effects_M1)+
geom_point(aes(factor, AME)) +
geom_errorbar(aes(x = factor, ymin = lower, ymax = upper)) +
geom_hline(yintercept = 0) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45))
Alternatively, I would like to know whether is is possible to integrate the confint command into a normal effects plot (see the one above) which will allow me to graph it together with the confidence intervals.
