1.2.0

What's Changed

• survey_prop() now uses proportions as the default, which should confidence interval improve coverage, but does mean results may slightly change (#141, #142, thanks @szimmer)
• New function survey_corr() calculates the correlation between 2 variables, (#150, #151, thanks @szimmer & @bschneidr)
• Added method for dplyr::reframe for srvyr's objects
• The function as_survey_rep() now accepts type="ACS" and type="successive-difference"
  for user-supplied replicate weights. (#153, thanks @bschneidr)

Full Changelog: v1.1.2...v1.2.0

v1.1.2

srvyr 1.1.2

• Fix for upcoming version of tidyselect (#147, thanks @hadley)

v1.1.1

srvyr 1.1.1

• Add function cur_svy_wts() to access the survey weights (#136, #139, thanks @ray-p144 and @bschneidr)
• Allow access to survey context functions like cur_svy() and cur_svy_wts() in mutate and filter (#138, #139, thanks @ray-p144 and @bschneidr)
• Improve behavior of interact() when using cascade()(#133, thanks @szimmer)
• Fix a bug with non-standard names of grouping variables (like 1234) in cascade (#132, thanks @szimmer)

v1.1.0

srvyr 1.1.0

• Uses the new quantile functions provided in version 4.1 of the survey package. The old survey quantile functions can be accessed with survey_old_quantile() and survey_old_median()
• Adds a new function interact that makes it easier to calculate proportions among interacted groups
• "Filering joins" (anti_join and semi_join) are now available for srvyr objects. You must put the tbl_svy object first. (#65, #120, @bschneidr)
• Auto-unpacking of data.frames works even inside of a named data.frame column (like one created by dplyr::across). (#129)
• Miscellaneous documentation improvements (#119, #126, #127)

1.0.1

srvyr 1.0.1

• survey_mean() with no x no longer errors when there are no grouping variables (#117)

srvyr 1.0.0

• summarize has been rearchitected,

  • main user facing improvements are:
    • dplyr::across() now works within it
    • dplyr functions like dplyr::cur_group(), dplyr::cur_group_id(), dplyr::cur_data()
      work in it (as well as new anlagous functions srvyr-specific cur_svy() and
      cur_svy_full())
  • The only known breaking change is:
    • objects in the summarize will refer to the output of summarize before the input.
      Meaning code that looks like this:

      dstrata %>% summarize(api99 = survey_mean(api99), api_diff = survey_mean(api00 - api99)) 

      will now error because it calculates the mean of api99 before using it inside of the
      calculation for api_diff. This behavior better matches dplyr's so will likely be
      kept.

• Support for group_map()/group_walk()/group_map_dfr(), group_split(),
  group_nest() and nest_by() were added for tbl_svy objects.

• Support drop_na from tidyr (#107).

• as_survey() and as_survey_() are now idempotent: given a srvyr survey object (a tbl_srv), they return it unchanged. If extra arguments are provided, they are ignored with a warning (#97, thanks @krivit).

• rename_with() now works with surveys (#96, thanks @krivit).

v0.4.0

srvyr (development version)

• Fix to ensure that ordered factors can be used as grouping variables or as inputs to survey_count and survey_tally (#92, thanks for reporting @szimmer & @walkerke & for fixing @bschneidr).

• Fix to ensure that numeric values can be used in grouping variables (#78 & #74, thanks for reporting @tzoltak & fix @bschneidr)

• Some improvements for dplyr 1.0 (#79) transmute() now works (thanks for reporting @caayala), summarise()'s .groups argument is respected, and multi-row returns to summarise() work. (Unfortunately the new across() function isn't quite supported in summarise() yet, it will hopefully come soon)

v0.3.10

srvyr 0.3.10

• Another fix for upcoming dplyr
• Fix in vignette for changes to vardpoor package

v0.3.9

srvyr 0.3.9

• Fix for upcoming version of dplyr (thanks @romainfrancois)

v0.3.8

• unweighted now evaluates in the right context and so will provide correct error
  when an incorrectly interpolated function is used (#70, thanks for reporting @tlmcmurry)

• filter_at works now, (#57, thanks for reporting @dcaseykc & helping @bschneidr).

• Fix for upcoming version of tibble (#72).

srvyr 0.3.7

srvyr 0.3.7

• filtering on grouped survey designs now works correctly (#54, thanks for reporting @dcaseykc)

• Added function pull (#63, thanks @dcaseykc)

• df parameter now set to be degrees of freedom of survey for quantiles and variance to match other
  functions.

• Updated tests to work with upcoming version of survey (#66).