Using syvtable in a R function

I would like to create a function which takes an x (character variable) in argument and put it into the svytable() function as part of the formula.

prop <- function(x){
  
  tab <- svytable(~x+group_treatment, dw)
  df <- as.data.frame(cprop(tab))
  df_bis <- reshape(df,idvar = "x",timevar = "group_treatment", direction = "wide")
  df_bis

}

prop("tranche_age")

When I run this code, I get the following error message : "Error in eval(predvars, data, env) : object ‘x’ not found
Called from: eval(predvars, data, env)".

Any idea to solve this issue?

Thanks,

Chloe

>Solution :

You could use reformulate to dynamically create the formula based on the column name passed to your function. Additonally you have to use idvar = x instead of idvar = "x" in reshape().

Using the default example data from ?svytable:

library(survey)
library(questionr)

data(api)

dw <- svydesign(id=~dnum, weights=~pw, data=apiclus1, fpc=~fpc)

prop <- function(x) {
  tab <- svytable(reformulate(c(x, "stype")), dw)
  df <- as.data.frame(cprop(tab))
  df_bis <- reshape(df, idvar = x, timevar = "stype", direction = "wide")
  df_bis
}

prop("sch.wide")
#>   sch.wide     Freq.E    Freq.H Freq.M  Freq.All
#> 1       No   8.333333  21.42857     32  12.56831
#> 2      Yes  91.666667  78.57143     68  87.43169
#> 3    Total 100.000000 100.00000    100 100.00000