Hello, i have a data set with only categorical variables. I am interested to examine the factors that affect voters to vote & the factors that affect voters to vote particular candidates. I want to run two models:
1.) My dependent variable (DV) is a binary variable of vote/no vote
2.) My DV is a categorical variable that holds the name of the candidates
My independent variables are all categorical variables such as: age bins, sex, geographic location, annual income bins and more.
I was thinking to use a logistic regression for the first model and a multinomial logistic regressionvfor the second one but i am not sure if that's the best option. What do you think?
I would recommend factorial analysis models or path analysis ( explicative or predictive designs SEM).
You may be interested in this project
https://www.researchgate.net/project/Modelling-interactions-analysis-of-large-tables-of-counts-using-a-Poisson-random-effects-model
a lot of our applications are about voting outcomes, it explicitly concentrates on interactions between the variables you have.
If I would be you, I would have followed the same models. Categorical variables, I feel are best shown by the help of logistic regression.
- Badges
- Science topic
- Similar topics
- Political Science