How to check the importance of Dummy variable in multiple regression?

#1
Hi All,

I am doing a Multiple regression with one non numeric independent variable. I changed this variable into a dummy variable which is now taking a value of "0" and "1".

After this, I did a Multiple regression in SPSS and find out that the significance value for this variable is greater than 0.05.
Moreover, I also did an F-Test in this model and a model without dummy variable to check which one is better. F-test also shows that adding this dummy variable is not adding any value in the model.

but, somehow I am not confident that this is correct. Is there any others means also to check whether this dummy variable has any effect on the model or not?
 
#2
Chow test can be used.

can you please specify what exactly do you mean by non-numeric independent variable and why did you chose a binary variable (0 and 1 only)? also, why are you uncomfortable if the tests are showing that dummy variable is not really needed? do you have special info (e.g. suggested by theory or prior research) about this variable which suggests that it should be important?
 
#3
I have profit earned in a casino as dependent variable and independent variables are customer entering, total games going on and so on. One such variable is day of operation, i.e. weekend or not. I have converted this variable into a dummy variable
weekend - 1
weekdays - 0
Now, my intuition is that, casino should make more money on weekends, but when i did regression, this variable is proving out to be non-significant.

I will try to do a chow test, as you suggest.
Thanks for the suggestion.