haven (version 2.5.0)

read_dta: Read and write Stata DTA files


Currently haven can read and write logical, integer, numeric, character and factors. See labelled() for how labelled variables in Stata are handled in R.

Character vectors will be stored as strL if any components are strl_threshold bytes or longer (and version >= 13); otherwise they will be stored as the appropriate str#.


  encoding = NULL,
  col_select = NULL,
  skip = 0,
  n_max = Inf,
  .name_repair = "unique"

read_stata( file, encoding = NULL, col_select = NULL, skip = 0, n_max = Inf, .name_repair = "unique" )

write_dta( data, path, version = 14, label = attr(data, "label"), strl_threshold = 2045 )



Either a path to a file, a connection, or literal data (either a single string or a raw vector).

Files ending in .gz, .bz2, .xz, or .zip will be automatically uncompressed. Files starting with http://, https://, ftp://, or ftps:// will be automatically downloaded. Remote gz files can also be automatically downloaded and decompressed.

Literal data is most useful for examples and tests. To be recognised as literal data, the input must be either wrapped with I(), be a string containing at least one new line, or be a vector containing at least one string with a new line.

Using a value of clipboard() will read from the system clipboard.


The character encoding used for the file. Generally, only needed for Stata 13 files and earlier. See Encoding section for details.


One or more selection expressions, like in dplyr::select(). Use c() or list() to use more than one expression. See ?dplyr::select for details on available selection options. Only the specified columns will be read from data_file.


Number of lines to skip before reading data.


Maximum number of lines to read.


Treatment of problematic column names:

  • "minimal": No name repair or checks, beyond basic existence,

  • "unique": Make sure names are unique and not empty,

  • "check_unique": (default value), no name repair, but check they are unique,

  • "universal": Make the names unique and syntactic

  • a function: apply custom name repair (e.g., .name_repair = make.names for names in the style of base R).

  • A purrr-style anonymous function, see rlang::as_function()

This argument is passed on as repair to vctrs::vec_as_names(). See there for more details on these terms and the strategies used to enforce them.


Data frame to write.


Path to a file where the data will be written.


File version to use. Supports versions 8-15.


Dataset label to use, or NULL. Defaults to the value stored in the "label" attribute of data. Must be <= 80 characters.


Any character vectors with a maximum length greater than strl_threshold bytes will be stored as a long string (strL) instead of a standard string (str#) variable if version >= 13. This defaults to 2045, the maximum length of str# variables. See the Stata long string documentation for more details.


A tibble, data frame variant with nice defaults.

Variable labels are stored in the "label" attribute of each variable. It is not printed on the console, but the RStudio viewer will show it.

If a dataset label is defined in Stata, it will stored in the "label" attribute of the tibble.

write_dta() returns the input data invisibly.

Character encoding

Prior to Stata 14, files did not declare a text encoding, and the default encoding differed across platforms. If encoding = NULL, haven assumes the encoding is windows-1252, the text encoding used by Stata on Windows. Unfortunately Stata on Mac and Linux use a different default encoding, "latin1". If you encounter an error such as "Unable to convert string to the requested encoding", try encoding = "latin1"

For Stata 14 and later, you should not need to manually specify encoding value unless the value was incorrectly recorded in the source file.


Run this code
path <- system.file("examples", "iris.dta", package = "haven")

tmp <- tempfile(fileext = ".dta")
write_dta(mtcars, tmp)
# }

Run the code above in your browser using DataCamp Workspace