logistic regression with 3-level-independent variable

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Pannie
Posts: 3
Joined: Thu Jan 16, 2014 2:25 pm

logistic regression with 3-level-independent variable

Postby Pannie » Wed Feb 26, 2014 8:00 pm

In my research I had a couple of dependent variables which didn't meet the assumptions of normality. Now my supervisor told me that I had to dichotomize these outcome variables and perform logistic regressions on these.

In all the examples of logistic regression on the internet, that I have seen so far, they use a couple of independent variables with just 2 levels. For example gender (1. female 2. male) or smoking (1. yes 2. no)

Now is the case that I have just one independent variable, with 3 levels:
Intervention:
1. No intervention
2. Memory training
3. Walking + Memory training

and I'm looking at the effect on (dichotomized) memory:
1. No recall
2. recall

Question: Can you perform logistic regression with one independent 3 level variable like you would with a 2 level independent variable?
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: logistic regression with 3-level-independent variable

Postby GerineL » Thu Feb 27, 2014 1:10 pm

look into multinominal regression for more than 2 outcomes.
however, you can probably just use chi square in crosstabs for your analyses, as you seem to have 2 nominal variables.

Who is online

Users browsing this forum: No registered users and 1 guest

cron