Write a data frame to SPSS, Stata or a TDA system file

These build TDA's internal data matrix from data and then hand it to TDA's writers. tda_write_spss produces a portable file (.por) or, with format = "sav", an SPSS system file. tda_write_stata produces a .dta for a chosen Stata release. tda_write_sys produces a TDA system file, which tda_read_sys reads back.

Usage

tda_write_spss(
  data,
  file,
  keep = NULL,
  drop = NULL,
  sort = NULL,
  format = c("portable", "sav"),
  options = list(),
  dir = tempfile("tda"),
  ...
)

tda_write_stata(
  data,
  file,
  keep = NULL,
  drop = NULL,
  sort = NULL,
  release = 10,
  options = list(),
  dir = tempfile("tda"),
  ...
)

tda_write_sys(data, file, options = list(), dir = tempfile("tda"), ...)

Arguments

data

a data frame.

file

path to write to.

keep, drop

variables to keep or drop, as a character vector. Give at most one of them.

sort

variables to sort the cases by, as a character vector.

format

for tda_write_spss: "portable" (the default, TDA's wspss) or "sav" (wspss1).

options

a named list of further TDA options, passed through.

dir

working directory.

...

passed to tda_run.

release

for tda_write_stata: the Stata release to write for – 4, 6, 7 or 10 (default).

Value

file, invisibly. attr(x, "run") carries the run.

Details

A TDA system file is TDA's format: it stores the data matrix together with the variable definitions, so reading one back restores the session's variables without repeating the nvar block.

Variable names

TDA will not read a variable name that does not begin with a capital – nvar(id = c1) is a syntax error to it, and the same parser handles keep, drop and sort. Columns are therefore capitalised on the way in, so id is written as Id. Give keep, drop and sort the names as they are in data; the capitalisation is applied for you.

See also

tda_read_spss, tda_read_sys

Examples

d <- data.frame(id = 1:5, y = c(2.5, 3, 1.5, 4, 2))
f <- tempfile(fileext = ".por")
tda_write_spss(d, f)
tda_read_spss(f, portable = TRUE)
#>   Id   Y
#> 1  1 2.5
#> 2  2 3.0
#> 3  3 1.5
#> 4  4 4.0
#> 5  5 2.0