# Very high standardized coefficient beta (binary logistic regression)

#### FlexTistic

##### New Member
I am comparing the influence of eleven independent variables on one binary dependent variable. I decided to use the standardized coefficient beta for the comparsion because some of the variables use different units of measurement.

One of my variables has a unstandardized coefficient of -0.0007 which I assume to be fine, because it is a \$ amount that goes from 0 to 1 Million and I also have some dependend dummy variables. But the standardized coefficent of that variable is over -7.5. As far as I know this indiciates multicollinearity but none of my variables have a VIF lower than 1 or higher than 1.7.

How do I know if I can just interpret the standardized coefficient or if I first need to do something else?

Sorry, if the question is confusing. I am a total beginner at statistics.

#### Attachments

• 31.1 KB Views: 3

#### hlsmith

##### Less is more. Stay pure. Stay poor.
I have never been able to process in my mind how a person would interpret a standardized categorical (e.g., > 2 groups) variable in a regression. Even a binary variable that was standardized is pretty darn weird to interpret. If you can't interpret them, do you really need to standardize them, the categorical variables that is.

#### FlexTistic

##### New Member
Thanks for the quick answer.

The only two categorial independent variables are binary variables and for these two the unstandardized coefficient is the same as the standardized one. So I shouldn't have any trouble interpreting these two.

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Perhaps we are talking about two different things, but a standardized binary variable would be 0 for its prevalence and a unit would be a std increase right. I didn't think it would have the same coefficient.

I didn't follow your original thoughts about collinearity?