You can have a look at methods of reduction of dimensionality, such as Primary Component Analysis.

Weighting by score maximization is able to combine several variables to have only one , indicative result of your search. I just used it and the result was modest.

thanks LAila, could you please guide me, how can i may do this. i have two variable which are maternal education and other is wealth status. both are categorical variables.

With pleasure @Atta muhammad asif. I propose you a discussion by meet-goole to show you how I proceeded in my work.

I think if your measuring scale for the variables that you want to add up is same then you can take simple mean score of them. You can try it. if measuring scale is different then look for reduction of dimensionality.

You may construct an index incorporating those two variables. And use that index as your variable in the model. Principle component analysis or Multiple correspondence analysis (depending on the nature of the variables) may be used to construct the index.