Region of origin, such as Africa, Asia etc is used as an instrumental variable for institutional quality which is usually assigned 1 or 0, 1 against that region to which that sample country belongs and 0 in rest of other region all samples belong. Estimation with such input creates problem. Anyone of research fellow may comment in this regard the optimal way to assign numeric weights to such variables please?

Note: same is applicable to legal origin e.g. British, french etc