haven 2.4.0

  2021/04/24

  haven

  Hadley Wickham

We’re delighted to announce the release of haven 2.4.0. haven allows you to read and write SAS, SPSS, and Stata data formats from R, thanks to the wonderful ReadStat C library written by Evan Miller.

You can install it from CRAN with:

install.packages("haven")

This blog post will show off the most important changes to the package; you can see a full list of changes in the release notes.

library(haven)

labelled_spss() and labelled()

labelled_spss() gains full vctrs support thanks to the hard work of Danny Smith. This means that labelled_spss() objects should now work seamlessly with dplyr 1.0.0, tidyr 1.0.0.

I’ve also made labelled() vectors are more permissive when concatenating. Now, output labels will be a combination of the left-hand and the right-hand side, and if there are duplicate labels, the left-hand side (first assigned) will win:

x1 <- labelled(1, labels = c(USA = 1))
x2 <- labelled(64, labels = c(NZ = 64))
c(x1, x2)
#> <labelled<double>[2]>
#> [1]  1 64
#> 
#> Labels:
#>  value label
#>      1   USA
#>     64    NZ

# It's now your responsibility to only combine things that make sense
x3 <- labelled(c(1, 2, 5, 3, 2), labels = c(Good = 5, Bad = 1))
c(x1, x3)
#> <labelled<double>[6]>
#> [1] 1 1 2 5 3 2
#> 
#> Labels:
#>  value label
#>      1   USA
#>      5  Good

Other improvements

  • Date-times are no longer converted to UTC. This should ensure that you see the same date-time in R and in Stata/SPSS/SAS. (But the underlying time point might be different because Stata/SPSS/SAS don’t appear to support time zones.)

  • Bundleed ReadStat has been updated to version 1.1.5 from 1.1.3 so includes ReadStat improvements in v1.1.5 and v1.1.4. Probably the biggest improvement is support for SAS-binary (aka Ross) compression.

  • write_*() now validates file and variable metadata with ReadStat, and validation failures now provide more details about the source of the problem (e.g. the column name), making it easier to track down issues.

Acknowledgements

A big thanks to everyone who helped make this release possible by asking questions, providing reprexes, writing code and more! @Ales-G, @atungate, @batpigandme, @bergen288, @bergsmat, @BernhardClemm, @bhaney22, @cimentadaj, @copernican, @DanChaltiel, @DavidLukeThiessen, @deschen1, @drag05, @drevanzyl, @dsteuer, @dswpg, @dusadrian, @elfatherbrown, @gorcha, @gowerc, @hadley, @hhchang0210, @hjvdwijk, @iamforerunner, @j-sirgo, @jacciz, @jackobailey, @jaydennord, @jimhester, @jkhanson1970, @kambanane, @krlmlr, @kwainfan, @larmarange, @lionel-, @MartinLBarron, @oliverbock, @ookiiwani, @peterolejua, @realrbird, @resuf, @rpruim, @rubenarslan, @sclewis23, @shannonpileggi, @sjkiss, @toerpe, @tslumley, @xlejx-rodsxn, @xmatic, and @zahlenzauber.

Contents