I have a binary variable in a dataframe , I want to assign the label "no" to the value 0 and "yes" to the value 1 without the vector becoming categorical (if this occurs, I can not use the function svymean() ).
Does anyone know if this is possible?
This is not possible.
What you can do is to let R transform into a factor, and when to use svymean do so:
x <- c("sim", "sim", "não", "não")
svymean(x = as.numeric(x == "sim"), design = ?)
See that using as.numeric(x == "sim") drops a vector of zeros and ones:
[1] 1 1 0 0