Conceptual queries in modeling

Hi all,
I am from engineering background. I would require your help in certain conceptual questions in modeling. Your help would be greatly appreciated!

Following are my few questions...
(Pointers on these questions would help me , else you may direct me to any useful resources.)

1)If a variable which is important from business standpoint has a p-value of 0.5, then should it be considered in the model? I read it as 'Yes' in literature but wouldn't it make the model coefficients unstable?

2)Is it mandatory to standardize the variables before building a logistic regression model? If Yes, is there a commonly followed approach?

3)I am planning to develop a logistic regression to rate the employees as good or bad. The model includes variables such as his innovation score, #papers published, salary, Training cost, etc. First two are kind of assets to the company and the next two are kind of liabilities. Should I explicitly make the model understand this by considering the liabilities as negative values?

4)I have two independent variables in my LR model. Var1 has levels 'A' and 'B'. Var2 has levels 'X' and 'Y'. Of the entire dataset, there are 30% observations with Var1 as 'A' and Var2 as 'X', 35% observations with Var1 as 'A' and Var2 as 'Y', 30% observations with Var1 as 'B' and Var2 as 'X', 5% observations with Var1 as 'B' and Var2 as 'Y'. The number of observations with Var1 as 'B' and Var2 as 'Y' are far too less compared to other combinations. Is this skewness in data going to affect my results? If so, how should I rectify this?

Thanks a lot in advance! :yup:


You do not need to standardise all your variables before running a logistic regression model.

Regarding liabilities variable, it is hard to answer this without knowing more about how the variable is measured, but you do not need to make adjustment to how the measure is varied in relation to other independent variables. As long as you know how it was measured i.e. 1 to 10, then you can interpret the output.

Regarding skewness - this is not unexpected with the categorical type of IV you are using.
Many people do not worry about skewness is their logistic regression.
Hello Evelyn,
Thanks a lot for the response!

My question regarding the standardization is that, wouldn't the coefficients get affected due to the scale of variables? Doesn't this make standardization mandatory? Let me have your thoughts.

Regarding liabilities, If I consider salary which is a continuous variable, then I should not be treating it any different than any other continuous variable in my model. Am I right?