Loop returns cumulative result, not discrete

Question

I'm trying to extract information from a several data files -- specifically, how many complete records exist in each file.

Here's what I've written:

complete <- function(directory, id=1:332) {
files_senscomp <- list.files(directory, full.names=TRUE)[id]
pre_dat <- data.frame()
full_dat <- data.frame()

for (i in seq(files_senscomp)) {
    pre_dat <- rbind(pre_dat, read.csv(files_senscomp[i]))
    nobs <- sum(complete.cases(pre_dat))
    id <- i
    full_dat <- rbind(full_dat,data.frame(id,nobs))
}
full_dat
}

What it returns, though, is cumulative. And the IDs are incorrect. Here's the function in action and the result:

> complete("specdata", 40:45)
  id nobs
1  1   21
2  2  248
3  3  308
4  4  382
5  5  665
6  6 1089

Why does this not return the IDs 40-45, along with a "nobs" result for an individual file rather than all of the files combined to that point?

Answer 1

This does it:

for (i in seq(files_senscomp)) {
    pre_dat <- read.csv(files_senscomp[i])  ## no `rbind`
    nobs <- sum(complete.cases(pre_dat))
    ID <- id[i]  ## `id` is your function argument, taking `40:45`
    full_dat <- rbind(full_dat,data.frame(id = ID, nobs = nobs))
    }