CRAN status Build status CRAN Downloads

Package to read and write all Stata file formats (version 17 and older) into a R data.frame. The dta file format versions 102 to 119 are supported.

The function read.dta from the foreign package imports only dta files from Stata versions <= 12. Due to the different structure and features of dta 117 files, we wrote a new file reader in Rcpp.

Additionally the package supports many features of the Stata dta format like label sets in different languages (?set.lang) or business calendars (?as.caldays).


The package is hosted on CRAN.



dat <- read.dta13("path to file.dta")
save.dta13(dat, file="newfile.dta")

Development Version

To install the current release from github you need the platform specific build tools. On Windows a current installation of Rtools is necessary, while OS X users need to install Xcode.

# install.packages("devtools")
devtools::install_github("sjewo/readstata13", ref="0.10.0")

Older Versions of devtools require a username option:

install_github("readstata13", username="sjewo", ref="0.10.0")

To install the current development version from github:

devtools::install_github("sjewo/readstata13", ref="testing")

Changelog and Features

Version Changes
0.10.1 fix writing “NA”, NA_character_ values
fix writing of STRLs in big endian systems
0.10.0 fix for reading/writing of format 119
fix sortlist attribute for dta format 119
fix compress option. In the past, unwanted conversions to integer type could occur.
fix encoding issues in variable and data labels
fix build on FreeBSD
new feature: improved handling of time and date formats
new feature: collect warnings from read.dta13
0.9.2 Fix Build on MacOS X
0.9.1 Allow reading only pre-selected variables
Experimental support for format 119
Improvements to partial reading. Idea by Kevin Jin
Export of binary data from dta-files
new function get.label.tables() to show all Stata label sets
Fix check for duplicate labels and in set.lang()
0.9.0 Generate unique factor labels to prevent errors in factor definition
check interrupt for long read. Patch by Giovanni Righi
Updates to notes, roxygen and register
Fixed size of character length. Bug reported by Yiming (Paul) Li
Fix saving characters containing missings. Bug reported by Eivind H. Olsen
Adjustments to convert.underscore. Patch by luke-m-olson
Allow partial reading of selected rows
0.8.5 Fix errors on big-endians systems
0.8.4 Fix valgrind errors. converting from dta.write to writestr
Fix for empty data label
Make replace.strl default
0.8.3 Restrict length of varnames to 32 chars for compatibility with Stata 14
Add many function tests
Avoid converting of double to floats while writing compressed files
0.8.2 Save NA values in character vector as empty string
Convert.underscore=T will convert all non-literal characters to underscores
Fix saving of Dates
Save with convert.factors by default
Test for NaN and inf values while writing missing values and replace with NA
Remove message about saving factors
0.8.1 Convert non-integer variables to factors (nonint.factors=T)
Handle large datasets
Working with strL variables is now a lot faster
<0.8.1 Reading data files from disk or url and create a data.frame
Saving dta files to disk - most features of the dta file format are supported
Assign variable names
Read the new strL strings and save them as attribute
Convert stata label to factors and save them as attribute
Read some meta data (timestamp, dataset label, formats,…)
Convert strings to system encoding
Handle different NA values
Handle multiple label languages
Convert dates
Reading business calendar files

readstata13 and foreign

Most attributes of the resulting data.frame are largely similar to the data.frames produced by foreign. Since newer Stata files require some additional attributes, the results of all.equal() and identical() will be FALSE for data.frames read by foreign::read.dta and read.dta13(). Otherwise, the data.frames produced by both functions are identical.


# with factors
r12 <- read.dta("")
r13 <- read.dta13("")

all.equal(r12, r13, check.attributes = FALSE)

# without factors
r12 <- read.dta("", 
                convert.factors = FALSE)
r13 <- read.dta13("", 
                  convert.factors = FALSE)

all.equal(r12, r13, check.attributes = FALSE)


Marvin Garbuszus (JanMarvin) and Sebastian Jeworutzki (sjewo)