I need to determine the increasing population impact on soil fertility in my district in Pakistan. I will take samples from different locations with apparently differently population. Is it correct if I consider number of families living per unit area (e.g. per square km etc)?