Generating a numeric variable from string variable

[Kareena Satia]

I currently have a variable called "Residence" which takes on two values: Urban and Rural

I'm generating a new variable called "Rural" which will take on values 1 for rural and 0 for urban. I'm creating this new variable based on this variable "Residence".

With the gen command, I'm facing problems of "type mismatch"

What should I do?

[Roman Mostazir]

I currently have a variable called "Residence" which takes on two values: Urban and Rural

What should I do?

The very first thing you could have done is to show us the Stata commands you have used. This is written in the FAQ section and we/forum users are expected to read it. There are many ways you can achieve what you want. One way is:

gen Rural = cond(Residence=="Rural",1,0)

[Kareena Satia]

Shows up as a "type mismatch"

[Wouter Wakker]

It looks like your Residence variable is in fact not a string variable but numeric, perhaps with value labels. Have a look at that and if the problem persists come back with a reproducible illustration of your problem, using a representative example of your data (make sure to use dataex) and the commands you used.

[Nick Cox]

So, you already have a numeric variable Residence, presumably with value labels.

tab Residence

tab Residence, nola 

will show it with and without labels. The code to get Rural as an indicator will depend on those results. It could even be that your variable Residence is already an indicator for being rural.

Roman Mostazir referred to the FAQ Advice that posters are asked to read first, which includes the advice to give a data example. https://www.statalist.org/forums/help#stata explains. An example in #1 would have allowed a precise and correct answer in #2.