Extract attributes from objects in a list and write them into a dataframe

March 30, 2023

I have a list of objects (outputs of survey::svyciprop) and I am trying to extract their attributes to create a data frame with all of the results.

#example of an object
props_and_cis[[1]]
                2.5% 97.5%
var1     0.932 0.826  0.98

I have figured out how to extract each item I want in a separate column:

var <- attr(props_and_cis[[1]],"names")
prop <- as.vector(props_and_cis[[1]])
ci_lower <- attr(props_and_cis[[1]], "ci")[1]
ci_upper <- attr(props_and_cis[[1]], "ci")[2]

I would like to iterate through each object in my list props_and_cis and write the extracted content into a dataframe, for example:

tribble(
  ~variable, ~prop, ~ci_lower, ~ci_upper,
  var,prop,ci_lower,ci_upper
)

But I can’t seem to make it work. Can someone help?

ETA:

> dput(props_and_cis[[1]])
structure(c(var1 = 0.932403111115339), var = structure(0.00119910004765771, dim = c(1L, 
1L), dimnames = list("as.numeric(var1)", "as.numeric(var1)")), ci = c(`2.5%` = 0.825647967272783, 
`97.5%` = 0.975715067477937), class = "svyciprop")

>Solution :

Write a function to extract the wanted data.

extract_attr <- function(x) {
  v <- attr(x, "names")
  ci <- attr(x, "ci")
  y <- cbind(data.frame(var = v, prop = c(x)), as.data.frame(t(ci)))
  row.names(y) <- NULL
  y
}

extract_attr(props_and_cis[[1]])
#>    var      prop     2.5%     97.5%
#> 1 var1 0.9324031 0.825648 0.9757151

^{Created on 2023-03-30 with reprex v2.0.2}

Then lapply the function to your list members and rbind the result to get one data.frame. In the example below i repeat the posted data example, all list members are equal to one an other.

res <- lapply(props_and_cis, extract_attr)
do.call(rbind, res)
#>    var      prop     2.5%     97.5%
#> 1 var1 0.9324031 0.825648 0.9757151
#> 2 var1 0.9324031 0.825648 0.9757151
#> 3 var1 0.9324031 0.825648 0.9757151

^{Created on 2023-03-30 with reprex v2.0.2}

Data

posted <-   structure(c(var1 = 0.932403111115339), 
                      var = structure(0.00119910004765771, dim = c(1L, 1L),
                                      dimnames = list("as.numeric(var1)", "as.numeric(var1)")),
                      ci = c(`2.5%` = 0.825647967272783, `97.5%` = 0.975715067477937), 
                      class = "svyciprop")

props_and_cis <- list(posted, posted, posted)

^{Created on 2023-03-30 with reprex v2.0.2}