questionr 0.7.7
- Fix error in
iorder when reordering a numeric variable with forcats (#131, @bassam-abulnoor)
- Fix error in
irec when recoding a numeric variable with forcats (#131, @bassam-abulnoor)
questionr 0.7.6
- Better tidyverse styling in irec and iorder (#125, @larmarange)
- include.lowest default value is now TRUE in icut() (#129, @nalimilan)
- Add
rp2018 dataset
- Remove
rp99 datset
questionr 0.7.5
- Fix
NA values in weights and normwt=TRUE in wtd.table
- bug fix in
describe() (#126)
questionr 0.7.4
- Add
useNA and deprecate na.rm and na.show arguments to wtd.table (thanks @nalimilan)
- Add
fct_relevel method to iorder (thanks @nalimilan)
- New function
ggsurvey() for easy plots with survey objects and ggplot2 (#123, @larmarange)
questionr 0.7.3
- Remove
dplyr::recode method from irec as it is in questioning lifecycle
- Fix
forcats and dplyr detection in irec (thanks @matthias-studer)
- Fix R.cache blocking message in
irec, icut or iorder (thanks @matthias-studer)
- Fix name conflicts when
irec, icut or iorder are called with a data frame with the same name as a function (thanks @nalimilan)
- Sort level names in
irec and iorder instead of relying on unique (thanks @nalimilan)
questionr 0.7.2
look_for() and lookfor() are now simply imported and reexported from labelled (#111, @larmarange)
questionr 0.7.1
fertility and fecondite datasets imported with haven 2.3.0 (thanks @larmarange)- Fix irec fct_recode code when there is a space in variable name
- Remove obsolete
wtd.var function. Use Hmisc::wtd.var instead.
- Improve styling of code generated by iorder, irec and icut
- Fix error when both “NA” and NA in a vector passed to
freq
- Use
dplyr::recode_factor instead of using factor() (thanks @larmarange)
questionr 0.7.0
- Compatibility with
labelled 2.0.0
questionr 0.6.3
- Make
rprop, cprop and prop compatible with janitor::tabyl for pipeline integration
- Replace
R2HTML::HTML with knitr::kable in clipcopy
- Fix incorrect NAs percentage in
describe (thanks @gdutz)
- Add new tabs() function (thanks @rdrr1990)
questionr 0.6.2
- Add
exclude argument to wtd.table (thanks @pgtpg)
- Make
clipcopy work with tibbles
- Bugfix : missing rownames in
iorder verification table
- Generate a
pkgdown package documentation at https://juba.github.io/questionr/
- Fix incompatibility between
cum, sort and NAs in freq (thanks @scoavoux)
- Bugfix : error when recoding a numeric variable with
forcats in irec
questionr 0.6.1
- New “Recoding addins” vignette
- Add support for
forcats::fct_recode in irec
- Add support for
dplyr::recode in irec
- “Variable cutting” addin entry renamed to “Numeric range dividing”
- Bugfix : conflict between
useNA and exclude in freq (thanks @scoavoux)
- Bugfix : Fix missing rownames in icut table results
questionr 0.6.0
- New
na.rm argument to cross.multi.table. Use na.rm = FALSE to display NA level in crossvar
- New
rp2012 dataset
- New
fertility dataset (@larmarange)
- New
ltabs function, for labelled variables cross-tabulation (@larmarange)
describe, lookfor and freq are harmonized and now work with labelled variables (@larmarange)- Integration with the
labelled package (@larmarange)
freq added to describe (@larmarange)- More detailed
lookfor results (@larmarange)
questionr 0.5.0
irec, icut and iorder have been converted to RStudio addins. They now work both with vectors and data frames.- Bugfix : handle regexp special chars in variable or split character in multi.split (thanks @markriseley)
irec allows to select the type of output : character, factor or numeric (thanks @larmarange)irec now works with numeric variables- Bugfix : name conflicts with global environment objects (thanks @scoavoux)
irec now trims leading and trailing whitespaces in inputted values- Minimal recoding style by default in
irec
- Fix false positives in multi.split (thanks @markriseley)
- New fecondite labelled dataset (thanks @larmarange)
- New happy dataset (thanks @briatte)
- cross.multi.table() now accept a
n argument to display the total number of observations by row or column (when freq=TRUE).
questionr 0.4.3
- cross.multi.table() now accept a
tfreq argument to display row percentages based on the (potentially weighted) contingency table of respondants.
- Fix : i* interactive functions now work with data.table and dplyr’s tbl_df
questionr 0.4.2
- cross.multi.table() now accept a
freq argument to display column percentages based on the (potentially weighted) contingency table of crossvar on respondants.
- multi.table() now accept a
freq argument to compute percentages based on (potentially weighted) number of repondants. Note that freq is set to TRUE by default
- Fix : wrong HTML() call in clipcopy()
- Refactoring : icut, irec and iorder are now shinyApps functions
- Translation : icut, irec and iorder interfaces are now translated in french
questionr 0.4.1
- Bugfix : compatibility with shiny 0.10
questionr 0.4
- New functions from Joseph Larmarange : duplicated2, na.rm, rm.unused.levels
- Default lookfor
keywords argument changed to "" (displays all variables)
- New argument
n to cprop and rprop to display the number of observations per row/column
- “Ensemble” and “Total” strings in cprop and rprop are now localized
- Bugfixes on irec : encoding on windows and empty strings in factor levels
freq now displays, by default, a column of percentage based on non-missing values
questionr 0.3.1
- Bugfix : weights handling in cross.multi.table
- Bugfix : NA values in icut
questionr 0.3
- New functions from Joseph Larmarange : addNAstr and odds.ratios
- copy renamed to clipcopy to avoid name collisions with data.table
- New function : icut, interactive shiny interface for
cut
- New function : iorder, interactive shiny interface for ordering the levels of a factor
- New function : irec, interactive shiny interface for recoding a variable
questionr 0.2
- New functions : describe, freq.na, qload, qscan, recode.na
- Finish transition from rgrs
questionr 0.1