Categorical Variables in Correlation Matrices

I have a matrix where several fields (Job Title, Grade, and Location) have more than one value. I have separated them out into their respective columns indicating a 1 if they are this value, and a 0 if they are not.

While running a correlation matrix with this format, one of these individual Job Title fields (Managers) of the categorical variable Job Title had an r indicator greater than .9. This is a single job title that is highly correlated with one Location. Both Job Title and Location are categorical variables. Does this single instance of correlation indicate multicollinearity throughout the entire Job Title field as well as the Location?

I know I can't remove just a single job title, but that I would instead have to remove the entire Job Title variable, or the same for Location.

Original Fields:

Job Titles Location
Manager Loc1
Supervis Loc3
Laborer Loc2
Laborer Loc3
Supervis Loc1
Laborer Loc2
Manager Loc3
Supervis Loc1
Laborer Loc2

New Fields:
Manager Supervis Laborer Loc1 Loc2 loc3
1 0 0 1 0 0
0 1 0 0 0 1
0 0 1 0 1 0
Etc... etc... for all instances. This doesn't show well in the forum, but you get my point I hope.

So now lets say Managers is highly correlated with Location3. Should we then remove the entire field of Job Titles or Locations since it's just ONE field TREATED as an independent variable, but really it's just PART of one variable.

Thanks for your help.