I am studying the driving's forces of crime in a particular city.
I have some independent values, e.g. percentage of the head of the household with low education per zone, the percentage of unemployed people per zone [...]
The percentage is related to the total population in each zone, and not from the total population of the whole city. I think if I use the absolute values it will give me some correlation (multicollinearity) among independent variable since that zone with more population tends to have also a higher number of people with low education, unemployed people and so on.
For example, instead of using the absolute value of the number of people with low education in a particular zone, maybe I should divide that value to the total population of that specific zone. This might help to avoid that such zones with higher population will not necessarily have a higher number of people with low level of education. Because a have difference in the size of zones and consequently the difference in their population.