I have a set of survey data, where I have the "Statefip code" for the US state an individual lives in. I want to create two binary variables:
1) NorthernStates: which has a value of 1 if they live in a "northern" state, and 0 otherwise
2) SouthernStates: which has a value of 1 if they live in a "southern state, and 0 otherwise
The definition I'm using for Southern state is when Statefip equals 1, 5, 12, 13, 22, 28, 37, 45, 47, 48, 51. These numbers correspond to Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, Virginia.
The definition I'm using for Northern state is when Statefip equals 9, 10, 17, 18, 23, 24, 25, 26, 33, 34, 36, 39, 42, 44, 50, 55. These numbers correspond to delaware, Illinois, Indiana, Maine, Maryland, Massachusetts, Michigan, New Hampshire, New Jersey, New York, Ohio, Pennsylvania, Rhode Island, Vermont, Wisconsin.
So for a west-coast or midwestern state (for example, California where Statefip==6) would equal 0 for both Southern and Northern variables.
Does anyone have a more elegant way to create these variables than me manually creating dummies for each state, and then adding them up to create the Southern and Northern states.
1) NorthernStates: which has a value of 1 if they live in a "northern" state, and 0 otherwise
2) SouthernStates: which has a value of 1 if they live in a "southern state, and 0 otherwise
| ID | Statefip | Northern | Southern |
| 234 | 1 | 0 | 1 |
| 546 | 12 | 0 | 1 |
| 876 | 18 | 1 | 0 |
| 432 | 33 | 1 | 0 |
| 156 | 6 | 0 | 0 |
The definition I'm using for Southern state is when Statefip equals 1, 5, 12, 13, 22, 28, 37, 45, 47, 48, 51. These numbers correspond to Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, Virginia.
The definition I'm using for Northern state is when Statefip equals 9, 10, 17, 18, 23, 24, 25, 26, 33, 34, 36, 39, 42, 44, 50, 55. These numbers correspond to delaware, Illinois, Indiana, Maine, Maryland, Massachusetts, Michigan, New Hampshire, New Jersey, New York, Ohio, Pennsylvania, Rhode Island, Vermont, Wisconsin.
So for a west-coast or midwestern state (for example, California where Statefip==6) would equal 0 for both Southern and Northern variables.
Does anyone have a more elegant way to create these variables than me manually creating dummies for each state, and then adding them up to create the Southern and Northern states.