Skip to contents

Read an SPSS data file

Source: R/data.R
tda_read_spss.Rd

tda_read_spss reads an SPSS system file (.sav, TDA's rspss1) or a portable file (.por, TDA's rspss) and returns its data as a data frame. The format is taken from the file extension unless portable says otherwise.

Usage

tda_read_spss(
  file,
  portable = grepl("\\.por$", file, ignore.case = TRUE),
  options = list(),
  dir = tempfile("tda"),
  ...
)

Arguments

file

path to a .sav or .por file.

portable

read as a portable (.por) file. Defaults to whether file ends in .por.

options

a named list of further TDA options, passed through.

dir

working directory.

...

passed to tda_run.

Value

A data frame, with attr(x, "run") carrying the run.

Details

These two commands were in TDA all along but had no wrapper, and no test: the only case in the suite fed rspss1 a text file and pinned its refusal. Both are now checked against real files written by readspss, and both round-trip their values exactly.

String columns do come through: rspss/rspss1 create real TDA string variables (type 1) and pdata prints them, so a character column arrives as text rather than a code. This is one of the few ways to get a string variable into TDA at all – tda_run refuses a character column in a data frame, because TDA's data-frame path has no type for it.

Two consequences worth knowing. TDA renames any variable whose name does not start with a capital (see Variable names must start with a capital), so a lowercase SPSS name comes back as VAR1, VAR2, ...; write the file with capitalised names if you need them preserved. And once a string variable exists, TDA's string operators work on it in a later nvar block of the same run – strlen(S), strsp(S) (the alphabetical sort position, matching R's rank), strv(S) and strvp(S,n,m).

For value labels and missing-value codes, use readspss directly: TDA keeps the data and the names, so a label or a user-missing code is lost on the way in, and this wrapper does not pretend otherwise.

See also

Other data: tda_read_xls(), tda_strings()

Examples

# readspss can write the fixture, so this runs where it is installed
if (requireNamespace("readspss", quietly = TRUE)) {
  f <- file.path(tempdir(), "example.sav")
  readspss::write.sav(data.frame(ID = 1:3, X = c(1.5, 2.5, 3.5),
                                 S = c("a", "bb", "ccc")), f)
  tda_read_spss(f)
}
#>   ID   X   S
#> 1  1 1.5   a
#> 2  2 2.5  bb
#> 3  3 3.5 ccc