Breaking
Arguments named group, at and group_by will be deprecated in future
releases. of easystats packages. Please use by instead. This affects
following functions in report:
report_participants()report_sample()Minor changes
report now supports reporting of Bayesian model comparison with variables of class brms::loo_compare.report now supports reporting of BayesFactor objects with variables of class BFBayesFactor.report_sample() now suggests valid column names for misspelled columns in the select, by, weights and exclude arguments.Bug fixes
report() for htest objects.New features
report_s() to report the interpretation of S- and p-values in an easy-to-understand
language.Major Changes
{report} to an MIT license.Minor changes
report now supports variables of class htest for the Chi2, Friedman test, Fisher's exact test, and Kruskal-Wallis.
report now supports variables of class Date, treating them like factors.
report now supports objects of class estimate_contrasts, from easystats'
modelbased::estimate_contrasts, outputting either the results in text form,
or as a table.
report_sample
ci, ci_method and ci_adjust arguments, to compute
confidence intervals for proportions of factor levels. Currently, two different
methods (Wald and Wilson) are available.group_by argument.group_by is allowed to be longer than 1).The print method for report_sample gains a layout argument, to print
tables either in "horizontal" or "vertical" layout.
Bug fixes
Fixed issue in report_participants, which did not print the "gender"
category for grouped output when that argument was written in lower-case.
Gender now also supports more alternate spellings, and age converts the
respective column to numeric.
Fixed printing issue for intercept-only models.
Hotfix for CRAN reverse dependency compatibility.
Breaking Changes
3.6.Minor changes
report_sample improvement
n argument to also optionally include sample size.total parameter was not respected.report_effectsize improvement
t.test (htest) objects, now support the type (one of c("d", "g"))
and rules (one of c"cohen1988", "sawilowsky2009", "gignac2016"))
arguments.BREAKING CHANGES
Minor changes
report_participants improvement (@rempsyc, #260)
Now correctly reports NA values as % missing
Adds support for country and race demographic information
BUG FIXES
report().BREAKING CHANGES
{report} and now live in
{datawizard}
package:Data wrangling helpers:
data_addprefix()
data_addsuffix()
data_findcols()
data_remove()
data_rename()
data_reorder()
Text formatting helpers:
format_text()
text_fullstop()
text_lastchar()
text_concatenate()
text_paste()
text_remove()
text_wrap()
MAJOR CHANGES
Reporting participant's sex/gender information has improved (thanks to @drfeinberg, #212)
Separated sex and gender into different searches/columns
Sex is reported % female, % male, % other, % missing if any cases are missing
Gender is reported % Women, % Men, % Non-Binary, % missing if any cases are missing
Age reports % missing if any cases are missing.
total column from
report_sample(), when grouping variable contained missing values.ivreg (ivreg).