Curious guy Posted March 21, 2015 Share Posted March 21, 2015 Hello Everyone So, I've come up the following problem - I have this data set with two variables - one is the age of a participant and the other one is a dichotomous variable that has to do with behavior. I want to know if the age affects the behaviour. Can I perform a t test, in which the age is the dependent variable? Obviously in theroy the age can only be the independent variable. If I perform the test the way I described it and find a significant difference (which I have), can I conclude that age affects behaviour, even though age was the dependent variable? Thanks! Ellies and MLHopeful 1 1 Link to comment Share on other sites More sharing options...
Igotnothin Posted March 21, 2015 Share Posted March 21, 2015 Hi there, I think it'd be a bit more natural to fit a logistic regression model with age as the predictor and the binary behavior variable as the outcome. Chances are it will agree with the t-test for whether mean age differs by those with the behavior and those without it. In other words, if one is significant, the other one probably will be too. But with the logistic regression you have what you see as the "cause" as the predictor and the "effect" as the outcome. Of course if the study design is observational you can't really determine causality. But still if you suspect age is what is affecting the behavior, then I think the logistic regression is the more natural statistical method. Curious guy 1 Link to comment Share on other sites More sharing options...
Curious guy Posted March 21, 2015 Author Share Posted March 21, 2015 Of course. I don't know why I didn't think about that. Thanks Igotnothin! MLHopeful and Igotnothin 1 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now