Variables in Logistic Regression Analysis

[danh32]

I am trying to come up with a model to predict a binary dependent solution. I want to incorporate ordinal independant variables as the predictors. Can I run a logistic regression analysis with ordinal variables, or do I need to recode them into dummy variables (0/1)? Or conversely, could I run an ordinal regression analysis using a binary variable? Let me know if my question is clear enough or if I need to clarify anything.

Thanks for the help!

[elnaz]

Dependent variable plays the role in selecting regression not independent variable,
it means that because your dependent variable is binary so you must use logistic regression.
please read the below explanations from SPSS help about logistic regression and ordinal regression:

Logistic regression is useful for situations in which you want to be able to predict the presence or absence of a characteristic or outcome based on values of a set of predictor variables.

Ordinal Regression allows you to model the dependence of a polytomous ordinal response on a set of predictors, which can be factors or covariates.

So you will lose some information and your accuracy in using Logistic regression instead of Ordinal Regression and its wrong to use Ordinal Regression instead of Logistic regression because your response in logistic regression is as nominal variable not ordinal.

[Link]

Are you referring to using an independent variable that is ordinal, with more than two values? If so, whether or not you use dummy variables depends. Are you willing to accept that the difference in effect between, say the first and the second values is the same as the difference in effect between, say the fourth and the fifth values? Its very unlikely that we ever except this, and so I would recommend dummy variables.

Also, depending on the question you want to answer, I'd say there are more methods than just logistic regression. Off the top of my head: Poisson, log-link binomial, and probit are a few. You should look into those too.