Changelog
Source:NEWS.md
ggstatsplot 0.12.4.9000
N.B. All statistical analysis in ggstatsplot is carried out in statsExpressions. Thus, to see changes related to statistical expressions, read the NEWS
for that package: https://indrajeetpatil.github.io/statsExpressions/news/index.html
MAJOR CHANGES
extract_stats()
returns a list of classggstatsplot_stats
which contains all the statistical summaries and expressions for a given plot.extract_stats()
,extract_subtitle()
,extract_caption()
now works out of the box for the grouped plots as well.
BUG FIXES
-
ggpiestats()
andggbarstats()
now respectratio()
argument for proportion tests run in case of two-way contingency tables (#818).
ggstatsplot 0.12.4
CRAN release: 2024-07-06
MAJOR CHANGES
- The feature to superimpose normality curve on a histogram (in
gghistostats()
) has been removed. This feature always felt like an ad hoc addition to the plot, and has nothing to do with the key statistical analysis in question (which is not about checking the normality of the distribution).
BUG FIXES
- Empty groups in factors are no longer dropped in
ggpiestats()
andggbarstats()
(#935).
ggstatsplot 0.12.3
CRAN release: 2024-04-06
MAJOR CHANGES
The cryptic but very useful parameter
k
has been renamed todigits
to improve its discoverability.To be consistent with other functions,
ggpiestats()
andggbarstats()
now default to two-sided alternative hypothesis.
ggstatsplot 0.12.1
CRAN release: 2023-09-20
MINOR CHANGES
Maintenance updates for changes in upstream dependencies.
ggbarstats()
gainssample.size.label.args
parameter to pass additional arguments toggplot2::geom_text()
.
ggstatsplot 0.12.0
CRAN release: 2023-08-07
BREAKING CHANGES
To be internally consistent, the
plot.type
argument has been removed fromggbetweenstats()
, since no such argument exists forggwithinstats()
. This argument was also redundant. Since removing a specific geom is straightforward using*.args
arguments. Examples for these two functions illustrate how.ggbetweenstats()
andggwithinstats()
retirepairwise.comparisons
argument since it was redundant. In order to turn off showing pairwise comparisons, you can now usepairwise.display = "none"
.
NEW FEATURES
-
ggbetweenstats()
getsboxplot.args
argument to pass additional arguments to the underlying geom function. This also fixes regression introduced in0.11.1
release where outlier points were displayed along with box plot.
ggstatsplot 0.11.1
CRAN release: 2023-04-14
BREAKING CHANGES
- The outlier tagging functionality in
ggbetweenstats()
andggwithinstats()
has been removed. It was too crude to be useful or reliable, and users should instead prefer more informative methods (e.g.performance::check_outliers()
).
MINOR CHANGES
- Fix failures due to changes in parameters.
ggstatsplot 0.11.0
CRAN release: 2023-02-15
BREAKING CHANGES
- The minimum needed R version is now bumped to
R 4.1
because a crucial dependency (pbkrtest) requires this R version.
ggstatsplot 0.10.0
CRAN release: 2022-11-27
BREAKING CHANGES
- The
output
parameter for all functions has been removed. All functions now return only the plot, which itself contains all necessary details that were previously extracted using theoutput
argument. You can extract all the necessary details (including expressions containing statistical details) from a plot usingextract_stats()
function. There are two additional helpers to get expressions:extract_subtitle()
andextract_caption()
.
MAJOR CHANGES
xfill
andyfill
arguments forggscatterstats()
have been removed. You can specify all aesthetic modifications for side histograms in scatter plot usingxsidehistogram.args
andysidehistogram.args
arguments.Updates to changes made in the latest ggplot2 release (
3.4.0
).
ggstatsplot 0.9.5
CRAN release: 2022-10-16
MAJOR CHANGES
Due to changes to the underlying API of parameters, the
effsize
argument has been renamed toeffectsize.type
.Removes unnecessary re-exports of tidyverse operators.
ggstatsplot 0.9.2
CRAN release: 2022-05-21
MAJOR CHANGES
The
pairwise_comparions()
function implementation now lives in statsExpressions package, although it will continue to be exported from ggstatsplot package.The details about pairwise test for
ggbetweenstats()
andggwithinstats()
functions are now displayed as a label for the secondary axis. Previously, this information was displayed in the caption. Given that caption already contained Bayesian test details, it was becoming difficult to stack different expressions on top of each other. To avoid unnecessary code complexity and also to avoid crowded caption, this decision was made. Additionally, the pairwise test label has been slightly abbreviated, and so is the label for significance bars. This is done to not let the text overwhelm the numeric values, the latter being more important.
ggstatsplot 0.9.1
CRAN release: 2022-01-14
MAJOR CHANGES
- Moves
{PMCMRplus}
package from Imports to Suggests. So, if, as a user, you wish to use pairwise comparisons inggbetweenstats()
andggwithinstats()
, you will need to download this package.
ggstatsplot 0.9.0
CRAN release: 2021-10-19
NEW FEATURES
- The
pairwise_comparisons()
function for carrying out one-way pairwise comparisons has now moved in ggstatsplot from{pairwiseComparisons}
package.
BREAKING CHANGES
A number of effect size estimates and their confidence intervals have changed due to respective changes made in effectsize package version
0.5
release. For full details of these changes, see: https://easystats.github.io/effectsize/news/index.htmlFor the same reason, the effect size for one-way contingency table has changed from Cramer’s V to Pearson’s C.
MAJOR CHANGES
For plotting marginal distributions in
ggscatterstats
, ggstatsplot now relies on ggside package instead of ggExtra. This was done to remove a glaring inconsistency in the API. All functions in ggstatsplot producedggplot
objects and could be further modified withggplot2
functions, exceptggscatterstats
, which led to a lot of confusion among users (e.g. #28). This change gets rid of this inconsistency. But it comes at a cost: there is no moremarginal.type
argument that lets you change the type of marginal distribution graphic and histogram is the only possible option. Note that this is not a breaking change. Your past code will continue to work but it will now always produce a histogram instead of other marginal graphic you might have chosen.Minimum needed R version is now
4.0
.
MINOR CHANGES
Online vignette about
combine_plots
has been removed. In case you want to create a grid of plots, it is highly recommended that you usepatchwork
package directly and not this wrapper around it which is mostly useful with ggstatsplot plots.ggscatterstats
labeling arguments accept only unquoted inputs now, and not quoted or string inputs. Allowing this was a bad design choice in the past since most functions in ggstatsplot, inspired bytidyverse
, expect unquoted (x
) - and not quoted ("x"
) - arguments. So this function was the odd one out.Gets rid of
ipmisc
dependency.Removes
movies_wide
dataset, which was virtually identical tomovies_long
dataset and was not used anywhere in the package. Also removes the unusedVR_dilemma
dataset.
ggstatsplot 0.8.0
CRAN release: 2021-06-09
MAJOR CHANGES
There is finally a publication for ggstatsplot package! https://joss.theoj.org/papers/10.21105/joss.03167
The
ggcoefstats
function defaults toNULL
forxlab
andylab
arguments, which lets users change these labels if they wish to do so. Additionally, the x-axis label, if not specified, now defaults to"estimate"
. Whether this estimate corresponds to regression coefficient or effect size like partial eta-squared should be clear from the label itself.To reduce the dependency load,
ggcorrplot
moves fromImports
toSuggests
.The
bar.fill
argument ingghistostats
is retired in favor of the newbin.args
argument that can be used to pass aesthetic arguments toggplot2::stat_bin
.ggstatsplot.layer
argument has been retired. If the user chooses a certainggplot2
theme, it means they want that theme, and not ggstatsplot’s varnish on it. So the previous behavior was undesirable. This is a backward compatible change, so the plots should not look different.
MINOR CHANGES
The
pch
size forggcorrmat
has been increased to 14 (#579) to increase its visibility compared to the correlation value text.ggwithinstats
gainspoint.args
to changegeom_point
.Minor change to
ggcorrmat
legend title - content in parentheses is now shown outside of it.
ggstatsplot 0.7.2
CRAN release: 2021-04-12
MAJOR CHANGES
To reduce the dependency load,
ggExtra
moves fromImports
toSuggests
.All functions are more robust in the sense that when statistical analysis fails, they will return only the plots with no subtitles/captions. This helps avoid difficult-to-diagnose edge case failures when the primary functions are used in
grouped_
functions (e.g., #559). Theggpiestats
andggbarstats
functions always behaved this way, but the rest of the functions now also mimic this behavior.
ggstatsplot 0.7.1
CRAN release: 2021-03-11
MAJOR CHANGES
- Based on feedback from the users, the argument
title.prefix
is now removed. It led to redundant title prefixes across different facets of the plot. Given thatgrouped_
functions require users to setgrouping.var
, it is fair to assume what variable the levels in the title correspond to.
MINOR CHANGES
Adapts to changes made in
statsExpressions 1.0.0
.sample.size.label
argument is retired forggbetweenstats
,ggwithinstats
, andggbarstats
. I do not think it is ever a good idea to not do this. If the users wish to not display sample sizes, they can easily do this usingscale_*
functions fromggplot2
.In
ggpiestats
andggbarstats
, parametric proportion tests are now turned off whentype = "bayes"
.
ggstatsplot 0.7.0
CRAN release: 2021-02-19
BREAKING CHANGES
combine_plots
has been completely revised to rely not onpatchwork
, but onpatchwork
, to combine a list ofggplot
together. This was done to have a leaner syntax. With this revision, its vestigial twincombine_plots
is no longer needed and has been removed. This should not break any of the existing instances ofgrouped_
functions, although it will lead to changed graphical layouts. The only instance in which this change will lead to a breakage is when you specifiedlabels
argument. So, if you usedplotgrid.args = list(labels = "auto")
, you will now have to replace it withplotgrid.args = list(tag_level = "keep")
. You can also useannotation.args
(e.g.,annotation.args = list(tag_levels = "a")
to customize labels (this will create labels with patterna
,b
,c
, etc.). Another instance of breakage is if you had usedcombine_plots
function and provided individual plots to...
instead as alist
.To avoid confusion among users, the default trimming level for all functions is now changed from
tr = 0.1
totr = 0.2
(which is whatWRS2
defaults to).
MAJOR CHANGES
All robust tests in this package were based on trimmed means, except for correlation test. This has been changed: the robust correlation measure is now Winsorized correlation, which is based on trimming. Therefore, the
beta
argument has been replaced bytr
argument. This should result only in minor changes in correlation coefficient estimates.Using
annotate
instead ofgeom_label
had significantly slowed downgghistostats
andggdotplotstats
functions. This has been fixed.Removes the vestigial
notch
andnotchwidth
arguments forggbetweenstats
andggwithinstats
.All Bayesian expression templates are now explicit about the type of estimate being displayed.
For
gghistostats
andggdotplotstats
, the centrality measure labels used to be attached to the vertical line, but this occluded the underlying data. Now this label is instead shown on the topx
-axis. Note that this means that if you make any further changes to the resulting plot using theggplot2::scale_x_continuous
function, this label will likely disappear. Thecentrality.k
argument is retired in favor ofk
.
NEW FEATURES
More models supported in
ggcoefstats
:crr
,eglm
,elm
,varest
.ggbetweenstats
,ggwithinstats
,gghistostats
,ggdotplotstats
gain argumentcentrality.type
that can be used to specify which centrality parameter is to be displayed. So one can havetype = "robust"
and still show median as centrality parameter by choosingcentrality.type = "nonparametric"
.
ggstatsplot 0.6.8
CRAN release: 2021-01-19
MAJOR CHANGES
gghistostats
removesbar.measure
argument. The function now defaults to showing thecount
information on thex
-axis and theproportion
information on the duplicatedx
-axis.ggscatterstats
removesmethod
andmethod.args
arguments. It will no longer be possible to use this function to visualize data for when the model is not linear. It also retiresmargins
argument.For
ggbetweenstats
andggwithinstats
functions, the arguments of typemean.*
have all been replaced bycentrality.*
. This is because now these functions decide which central tendency measure to show depending on thetype
argument (mean for parametric, median for non-parametric, trimmed mean for robust, and MAP estimator for Bayes).Similarly,
gghistostats
andggdotplotstats
functions also decide which central tendency measure to show depending on thetype
argument (mean for parametric, median for non-parametric, trimmed mean for robust, and MAP estimator for Bayes). Therefore,centrality.parameter
argument has been removed. If you want to turn off displaying centrality measure, setcentrality.plotting = FALSE
.gghistostats
andggdotplotstats
functions remove the functionality to display a vertical line corresponding totest.value
. This feature was turned off by default in prior releases. Accordingly, all related arguments from these two functions have been removed.ggscatterstats
defaults todensigram
as the marginal distribution visualization.ggbetweenstats
andggwithinstats
now display the centrality tendency measure in such a way that the label doesn’t occlude any of the raw data points (#429).mean.ci
argument is retired forggbetweenstats
andggwithinstats
. Future ggstatsplot releases will be providing different centrality measures depending on thetype
argument and it is not guaranteed that all of them will have CIs available. So, for the sake of consistency, this argument is just going to be retired.
MINOR CHANGES
ggcorrmat
uses pretty formatting to display sample size information.ggcoefstats
now also displays degrees of freedom for chi-squared tests.Expects minor changes in some of the effect sizes and their confidence intervals due to changes in statsExpressions.
ggstatsplot 0.6.6
CRAN release: 2020-12-03
BREAKING CHANGES
ggcoefstats
no longer supportsexponentiate
argument. If it is specified, the user will have to themselves adjust the scales appropriately.-
ggcorrmat
defaults have changed significantly:As a matter of good practice, the p-values are adjusted by default for multiple comparisons.
The default matrix is upper type, and not the full matrix, which features many redundant comparisons and self-correlations diagonally.
Default text size for legend has been increased to 15 and background grid has been removed.
BUG FIXES
In the prior release, when the GitHub version of
BayesFactor
wasn’t present,ggwithinstats
just outright failed to run for ANOVA designs. This has been fixed.Setting
mean.path = FALSE
inggwithinstats
produced incorrect colors for points (#470). This bug was introduced in0.6.5
and is now fixed.If user had set
options(scipen = 999)
in their session, the p-value formatting forggpiestats
andggcoefstats
looked super-ugly (#478). This has been fixed.
MAJOR CHANGES
Drops
broomExtra
from dependencies. All regression modeling-related analysis now relies oneasystats
ecosystem.ggpiestats
andggbarstats
don’t support returning dataframes. See FAQ vignette on how to get these dataframes: https://indrajeetpatil.github.io/ggstatsplot/articles/web_only/faq.html#faq-1ggpiestats
andggbarstats
were not supposed to support returning Bayes Factor for paired contingency table analysis, which is not supported inBayesFactor
itself.ggcoefstats
defaults to displaying the intercept term. Also, when the degrees of freedom are not available fort
-statistic, they are displayed to beInf
, in keeping witheasystats
conventions.Instead of showing significance of p-values with APA’s asterisks conventions,
ggbarstats
now instead shows the actual p-values from one-sample proportion tests.
ggstatsplot 0.6.5
CRAN release: 2020-10-31
BREAKING CHANGES
-
ggpiestats
andggbarstats
no longer have the vestigial argumentsmain
andcondition
, which are superseded byx
andy
, respectively.
MAJOR CHANGES
For consistency and to reduce confusion, all Bayes Factor (irrespective of whether in the subtitle or caption) are always in favor of null over alternative (
BF01
).Retires centrality parameter tagging functionality of
ggscatterstats
. Although it was not the default, when turned on, it definitely created a cluttered plot.
ggstatsplot 0.6.1
CRAN release: 2020-10-06
MAJOR CHANGES
-
ggbetweenstats
andggwithinstats
functions now default topairwise.comparisons = TRUE
.
ggstatsplot 0.6.0
CRAN release: 2020-09-13
NEW FEATURES
More models supported in
ggcoefstats
:BFBayesFactor
,betamfx
,crq
,coxph.penal
,geeglm
,glht
,glmm
,lm_robust
,lqm
,lqmm
,manova
,maov
,margins
,negbinmfx
,logitmfx
,logitsf
,margins
,poissonmfx
,betaor
,negbinirr
,logitor
,metafor
,metaplus
,orm
,poissonirr
,semLm
,semLme
,vgam
.ggpiestats
gainslabel.repel
argument to cover contexts in which the labels might overlap. Setting it toTRUE
will minimize such an overlap.ggbetweenstats
andggwithinstats
gainggsignif.args
argument to make it easy to change aesthetics of the pairwise comparison geom.The subtitle and caption for Bayes Factor tests now also provide information about posterior estimates, when relevant.
MAJOR CHANGES
Removed unused
intent_morality
dataset.ggcoefstats
retirescaption.summary
argument. So, by default, the caption is going to contain as much information as it can and the users can then choose to modify the default caption usingggplot2
functions.
MINOR CHANGES
The argument
method
forggcorrmat
has been renamed tomatrix.method
, since it was confusing whether this method referred to correlation method.For both
ggpiestats
andggbarstats
, the count labels no longer includen =
in them as this was confusing since all labels hadn =
in them with no further explanation about how thisn
differed fromn
in the proportion test.No longer relies on
groupedstats
package.
ggstatsplot 0.5.0
CRAN release: 2020-05-30
BREAKING CHANGES
-
The
pairwise.annotation
argument forggbetweenstats
andggwithinstats
is deprecated. This was done because-Different fields have different schema for what significance levels asterisks represent.
The p-value labels also contain information about whether they are adjusted for multiple comparisons.
The
normality_message
andbartlett_message
helper functions have been removed. This is because model assumption checks don’t really fall under the purview of this package. There are excellent visualization tools out there for model assumption checks (ggResidpanel
,performance
,DHARMa
,olsrr
, etc.), which should be preferred over unhelpful messages with only p-values that these functions were printing. For what it’s worth, the functions where these messages were displayed (ggbetweenstats
orggwithinstats
) feature visualizations rich enough and defaults sensible enough that most of the time one can either assess these assumptions from the plots or need not worry about them.
MAJOR CHANGES
ggcoefstats
has been refactored to reflect thatbroomExtra::tidy_parameters
now defaults toparameters
package instead ofbroom
. It also loses the following vestigial arguments:p.adjust.method
andcoefficient.type
.Reverts aligning title and subtitle with the plot and not the axes, since it looked pretty ugly (esp.,
ggcoefstats
) and was causing problems for labels.factor.levels
(forggpiestats
) andlabels.legend
(forggbarstats
) are deprecated. If users would like to changes the names for factor levels, this should be done outside of ggstatsplot.The non-parametric post hoc test for between-subjects design has been changed from Dwass-Steel-Crichtlow-Fligner test to Dunn test.
ggstatsplot 0.4.0
CRAN release: 2020-04-15
BREAKING CHANGES
ggcorrmat
no longer returns matrices of correlation coefficients or other details. It now returns either a plot or a data frame and this can data frame can then be used to create matrices.ggbarstats
losesx.axis.orientation
argument. This argument was supposed to help avoid overlapping x-axis label, but nowggplot2 3.3.0
has a better way to handle this: https://www.tidyverse.org/blog/2020/03/ggplot2-3-3-0/#rewrite-of-axis-code
NEW FEATURES
More models supported in
ggcoefstats
:bayesx
,BBmm
,brmultinom
,lmerModLmerTest
,lrm
.Specifying
output = "proptest"
forggpiestats
andggbarstats
functions will now return a data frame containing results from proportion test.ggbetweenstats
andggwithinstats
will display pairwise comparisons even ifresults.subtitle
is set toFALSE
.ggcorrmat
supports computing Bayes Factors for Pearson’s r correlation.ggbetweenstats
andggwithinstats
now support pairwise comparisons for Bayes Factor test.
MAJOR CHANGES
For changes related to subtitle details, see changes made in new version of
statsExpressions 4.0.0
: https://CRAN.R-project.org/package=statsExpressions/news/news.htmlggbetweenstats
andggwithinstats
no longer print dataframes containing results from pairwise comparisons tests because this is too cluttering for the user’s console. The users are now instead advised to either extract this data frame usingggplot2::ggplot_build()
function or use thepairwiseComparisons::pairwise_comparisons()
function used in the background by ggstatsplot to carry out this analysis.Due to changes in one of the downstream dependencies, ggstatsplot now expects the minimum R version to be
3.6.0
.
MINOR CHANGES
ggcorrmat
now internally relies oncorrelation
for correlation analyses.ggbarstats
no longer displays"percent"
for Y-axis label as this was redundant information.Continuing the argument cleanup that began in
0.3.0
,ggcoefstats
gainspoint.args
argument instead of individualspoint.*
arguments.The subtitles are more explicit about the details of the test. For the same reason
stat.title
argument from all relevant functions is retired since this argument was supposed to be for entering some additional details about the test. Additionally, the plot titles and subtitles for some of the plots are aligned with the plot.ggcorrmat
legend, in case of missing values, shows mode - instead of median - for the distribution of sample pairs.-
The following vestigial arguments are retired:
caption.default
inggcorrmat
k.caption.summary
inggcoefstats
ggstatsplot 0.3.1
CRAN release: 2020-03-06
This is a hotfix release to correct some of the failing tests and other minor breakages resulting from the new release of ggplot2 3.3.0
.
ggstatsplot 0.3.0
CRAN release: 2020-03-01
BREAKING CHANGES
To give users more flexibility in terms of modifying the aesthetic defaults for all geoms
included in the ggstatsplot plots (each plot typically has multiple geoms), the package now uses a new form of syntax. Previously, each geom
had a separate argument to specify each aesthetic (e.g., geom_point
would get arguments like point.size
, point.color
, etc.), which resulted in functions with a massive number of arguments and was unsustainable in the long run. Instead, ggstatsplot functions now expect a list of such arguments for the respective geom (e.g., geom_point
will have point.args
argument where a list of arguments list(size = 5, color = "darkgreen", alpha = 0.8)
can be supplied).
All
grouped_
functions have been refactored to reduce the number of arguments. These functions now internally use the newcombine_plots
instead ofcombine_plots
. The additional arguments to primary functions can be provided through...
. These changes will not necessarily break the existing code but will lead to some minor graphical changes (e.g., if you were providinglabels
argument explicitly, it will be ignored).All functions lose the
return
argument, which was supposed to be alternative to enteroutput
. But this was just leading to more confusion on the user’s part. The biggest user-visible impact this is going to have is thatggcorrmat
will no longer be backward-compatible. The older scripts will still work but if thereturn
argument was anything except"plot"
, it will just be ignored.ggcorrmat
no longer hascorr.method
argument. To be consistent with rest of the functions in this package, the type of statistics should be specified usingtype
argument. Additional, it gains a new argumentggcorrplot.args
, which can be used to pass additional arguments to the underlying plotting function (ggcorrplot::ggcorrplot
).Both
gghistostats
andggdotplotstats
now use the following arguments to modifygeom
s corresponding to the lines and labels:test.value.line.args
,test.value.label.args
,centrality.line.args
,centrality.label.args
. This helps avoid specifying millions of arguments.Removes the vestigial
ggplot_converter
function.ggpiestats
andggbarstats
remove the following vestigial arguments:facet.wrap.name
,bias.correct
,bar.outline.color
. Thebar.proptest
andfacet.proptest
arguments were difficult to remember and confusing and are replaced by a commonproportion.test
argument. Additionally, the following arguments have all been removed and replaced bylabel
argument:slice.label
,bar.label
,data.label
. These plethora of options was a headache to remember.gghistostats
loses the following arguments:fill.gradient
,low.color
,high.color
. It made no sense to add a color gradient to this plot when the Y-axis already displayed the information about what the bar represented.ggscatterstats
loses the following arguments:palette
andpackage
. Since this function requires only two colors, it didn’t make much sense to use color palettes to specify this. They can be instead specified usingxfill
andyfill
. You can always usepaletteer::paletteer_d
to get a vector of color values and then provide values of your choosing toxfill
andyfill
.Removes sorting options in
ggbetweenstats
andggwithinstats
functions. This is something the users can easily do before entering the data in these functions.
MAJOR CHANGES
ggcorrmat
was never supposed to work with Kendall’s correlation coefficient but it accidentally did. This is no longer the case.ggstatsplot now has a logo, thanks to Sarah! :)
The default
theme_ggstatsplot
changes slightly. The biggest change is that the title and the subtitle for plots are now aligned to the left of the plot. This change also forced the legend forggpiestats
to be displayed on the right side of the plot rather than at the bottom.
ggstatsplot 0.2.0
CRAN release: 2020-02-03
BREAKING CHANGES
To have a more manageable length of function arguments, additional aesthetic specifications for any given geom can be provided via a dedicated
*.args
argument. For example, all aesthetic arguments forgeom_vline
can be provided viavline.args
, forgeom_errorbarh
viaerrorbar.args
, etc.ggstatsplot continues with its conscious uncoupling that started in
0.1.0
release: The following functions have now been moved to statsExpressions package:subtitle_meta_parametric
andbf_meta_message
and follow a more logical nomenclature. For the same reason,lm_effsize_ci
function is also no longer exported and lives in thegroupedstats
package.
MAJOR CHANGES
The summary caption no longer displays log-likelihood value because it tends to be not available for a number of regression model objects and so the caption was unnecessarily being skipped.
Supports robust and Bayes Factors for random-effects meta-analysis.
MINOR CHANGES
New dataset included:
bugs_wide
More models supported in
ggcoefstats
:cgam
,cgamm
,coxme
,cpglm
,cpglmm
,complmrob
,feis
,flexsurvreg
,glmx
,hurdle
,iv_robust
,mixor
,rqss
,truncreg
,vgam
.Removed vestigial arguments from
ggcorrmat
(e.g.,exact
,continuity
, etc.) andggpiestats
(bf.prior
,simulate.p.value
,B
, etc.).
ggstatsplot 0.1.4
CRAN release: 2019-12-18
BUG FIXES
-
ggbetweenstats
andggwithinstats
no longer produce error with variables with patternmean
(#336).
MAJOR CHANGES
-
pairwise_p
has been reintroduced as a number of users found it useful to call the function from ggstatsplot itself rather than usingpairwiseComparisons
package.
MINOR CHANGES
ggbetweenstats
andggwithinstats
use[
instead of(
to display confidence intervals. Additionally, denoted sample mean, but was confused with population mean by some users. So these functions instead display .More models supported in
ggcoefstats
:bmlm
,coeftest
Adapts to the new syntax provided in
paletteer
package.
ggstatsplot 0.1.3
CRAN release: 2019-11-21
MAJOR CHANGES
- To avoid excessive arguments to function, most arguments relevant for
ggrepel
inggcoefstats
function have been removed. The users can instead provide all such arguments in a list tostats.labels.args
argument.
BUG FIXES
ggbetweenstats
andggwithinstats
no longer produce incorrect label if the data frame already contains a variable namedn
(#317) or variables with patternmean
(#322).ggbetweenstats
andggwithinstats
mean labels respectk
argument (#331).
MINOR
ggcoefstats
now usesparameters::p_value
instead ofsjstats::p_value
, as requested by the maintainer of that package. This might lead to differences in p-values forlmer
models.More models supported in
ggcoefstats
:blavaan
,bracl
,brglm2
,glmc
,lavaan
,nlreg
,slm
,wbgee
.ggcoefstats
gainsonly.significant
argument to only display display stats labels for significant effects. This can be helpful when a large number of regression coefficients are to be displayed in a single plot.
ggstatsplot 0.1.2
CRAN release: 2019-09-17
MINOR
- Minor code refactoring that gets rid of the following dependencies:
magrittr
,ellipsis
,purrrlyr
.
MAJOR
- The p-value label now specifies whether the p-value displayed in
ggbetweenstats
andggwithinstats
pairwise comparisons were adjusted or not for multiple comparisons.
ggstatsplot 0.1.1
CRAN release: 2019-08-30
ANNOUNCEMENTS
ggstatsplot is undergoing conscious uncoupling whereby all the statistical processing functions that make stats subtitles are being moved to a new package called statsExpressions. This new package will act as a backend that handles all things statistical processing. This will not affect the end users of ggstatsplot unless you have been using the helper functions.
Additionally, multiple pairwise comparison tests are being moved to an independent package called pairwiseComparisons
.
This uncoupling is designed to achieve two things:
Make the code base of more manageable size in ggstatsplot, which will make package development a bit easier.
Make the workflow more customizable since now you can prepare your own plots and then use statsExpressions to display results in the plot rather than relying on ggstatsplot default plots which are heavily opinionated and not appealing to everyone.
BREAKING CHANGES
All helper functions
subtitle_*
andbf_*
have been moved to the new statsExpressions package.To be consistent with all the other
subtitle_
andbf_
functions,subtitle_contingency_tab
andbf_contingency_tab
now use the argumentsx
andy
instead ofmain
andcondition
.
MAJOR CHANGES
Major refactoring to reduce the codesize and to rely fully on
rlang
.There was confusion about what did the red point in
ggbetweenstats
andggbetweenstats
plots represents. Now the label also contains to highlight that what is being displayed is a mean value.To be consistent with the rest of the functions,
ggpiestats
andggbarstats
now uses the following aliases for arguments:x
formain
andy
forcondition
. This change is backward-compatible and should not pose any problems for scripts that usedmain
andcondition
arguments in these functions.Most subtitle expressions now report details about the design. In case of between-subjects design, this will be , while in case of repeated measures design, this will be .
pairwise.annotation
now defaults to"p.value"
rather than"asterisk"
forggbetweenstats
andggwithinstats
(and theirgrouped_
variants) functions. This was done because the asterisk conventions are not consistent across various scientific disciplines.
MINOR CHANGES
New dataset included:
bugs_long
, for repeated measures designs withNA
s present in the data.ggstatsplot now uses
rcompanion
to compute Spearman’s rho and Kendall’s W. Therefore,DescTools
is removed from dependencies.ggcoefstats
supports following objects:bglmerMod
,blmerMod
,lme
,mclogit
,mmclogit
,tobit
,wblm
.ggcoefstats
now respectsconf.int
. It internally always defaulted toconf.int = TRUE
inbroom::tidy
irrespective of what was specified by the user.It was painfully confusing for a lot of users what exactly the asterisks in each facet of
ggpiestats
signified. So instead nowggpiestats
displays more detailed results from a goodness of fit (gof) test. No such change is made forggbarstats
because there is no space to include more details above the bar.Removed
conf.method
andconf.type
arguments forggcoefstats
. Also,p.kr
argument removed becauseggcoefstats
will begin to rely onparameters
instead ofsjstats
package to compute p-values for some regression models.
ggstatsplot 0.0.12
CRAN release: 2019-07-12
BUG FIXES
- Bayes Factor in
ggwithinstats
caption, displayed by default, was incorrect. This has been fixed. This stemmed from a line of code which should have beenpaired = TRUE
, but was insteadpaired = FALSE
.
MAJOR CHANGES
- The effect size measure for Kruskal-Wallis test has been changed from the more obscure H-based eta-squared statistic to more common and interpretable epsilon-squared.
MINOR CHANGES
ggcoefstats
defaults tobf.message = TRUE
to be consistent with the rest of the functions in the package.ggcoefstats
supports the following class of objects:epi.2by2
,negbin
,emmGrid
,lmrob
,glmrob
,glmmPQL
,data.table
.bf_ttest
is introduced as a general function. The previously exportedbf_one_sample_ttest
andbf_two_sample_ttest
become its aliases.bf_meta_message
syntax changes to adapt to updates made tometaBMA
package (thanks to #259).
BREAKING CHANGES
The vestigial arguments
axis.text.x.margin.t
,axis.text.x.margin.r
,axis.text.x.margin.b
,axis.text.x.margin.l
forggcorrmat
have been removed. The margins can be adjusted usingggplot2::margin()
.gghistostats
no longer allowsdata
argument to beNULL
. This is to make this function’s syntax consistent with rest of the functions in this package (none of which allowdata
to beNULL
). This also removes confusion that arose for some users whendata
couldn’t beNULL
for itsgrouped_
cousin (grouped_gghistostats
).outlier_df
function is no longer exported since it was always meant to be an internal function and was accidently exported during initial release and was retained for a while for backward compatibility.
ggstatsplot 0.0.11
CRAN release: 2019-06-14
BREAKING CHANGES
Instead of having two separate functions that dealt with repeated measures (
subtitle_friedman_nonparametric
) and between-subjects (subtitle_kw_nonparametric
), a single functionsubtitle_anova_nonparametric
handles both of these designs with thepaired
argument determining which test is run.All functions that supported Bayes Factor analysis (
type = "bf"
) will only return BF value and the scale used. Previously, this was a mix of parametric statistics and BF, which was confusing and often times misleading since these two types of analyses relied on different tests.The default for
bf.message
has been changed fromFALSE
toTRUE
. This is to make the Bayes Factor analysis more visible to the user.
MAJOR CHANGES
-
ggscatterstats
returns only plot (without any statistical details) when the specified model is not linear (i.e., either whenmethod
argument is not"lm"
or whenformula
is noty ~ x
).
NEW FEATURES
New functions
ggwithinstats
(and itsgrouped_
variant) are introduced as a counterpart toggbetweenstats
to handle repeated measures designs.For repeated measures ANOVA,
subtitle_anova_nonparametric
now returns confidence intervals for Kendall’s W.All functions get
return
argument that can be used to return either"plot"
,"subtitle"
, or"caption"
. This makes it unnecessary to remember which subtitle function is to be used where. As a result, in the next release, all subtitle making functions will not be exported and are encouraged not be used either by other developers or by users.Both
subtitle_anova_robust
andsubtitle_anova_parametric
gain a new argumentpaired
to support repeated measures designs.ggcoefstats
can support following new model objects:drc
,mlm
.ggcoefstats
gainsbf.message
argument to display a caption containing results from Bayesian random-effects meta-analysis. It therefore gains a new dependency:metaBMA
.ggpiestats
andggcatstats
will now display Cramer’s V as effect size for one-sample proportion tests.All functions gain
stat.title
argument (NULL
by default) that can be used to prefix the subtitle with a string of interest. This is possibly useful for specifying the details of the statistical test.
MINOR CHANGES
pairwise_p()
function no longer outputsconf.low
andconf.high
columns when parametric post hoc tests are run. This is because these values were accurate only when no p-value adjustment was carried out.Instead of using the internal function
cor_test_ci
,ggscatterstats
instead usedSpearmanRho
function fromDescTools
package. This was done to reduce number of custom internal functions used to compute CIs for various effect sizes. ggstatsplot therefore gainsDescTools
as a dependency.The
sampling.plan
argument default forggbarstats
function has been changed from"indepMulti"
to"jointMulti"
to be consistent with its sister functionggpiestats
.
ggstatsplot 0.0.10
CRAN release: 2019-03-17
NEW FEATURES
ggcoefstats
can support following new model objects:rjags
.New
VR_dilemma
dataset for toying around with within-subjects design.subtitle_t_onesample
supports both Cohen’s d and Hedge’s g as effect sizes and also produces their confidence intervals. Additionally, non-central variants of these effect sizes are also supported. Thus,gghistostats
and itsgrouped_
variant gets two new arguments:effsize.type
,effsize.noncentral
.ggpiestats
used to display odds ratio as effect size for paired designs (McNemar test). But this was only working when the analysis was a 2 x 2 contingency table. It now instead displays Cohen’s G as effect size, which generalizes to any kind of design.
MINOR CHANGES
The internal function
outlier_df
to add a column specifying outlier status of any given data point is now exported.ggstatsplot previously relied on an internal function
chisq_v_ci
to compute confidence intervals for Cramer’s V using bootstrapping but it was pretty slow. It now instead relies onrcompanion
package to compute confidence intervals for V. ggstatsplot, therefore, gains a new dependency.subtitle_mann_nonparametric
andsubtitle_t_onesample
now computes effect size r and its confidence intervals as (with the help ofrcompanion
package), instead of using Spearman correlation.
ggstatsplot 0.0.9
CRAN release: 2019-02-18
BREAKING CHANGES
-
subtitle_t_onesample
no longer hasdata
as the optional argument. This was done to be consistent with other subtitle helper functions.
NEW FEATURES
New function
ggbarstats
(and itsgrouped_
variant) introduced for making bar charts (thanks to #78).ggcoefstats
also displays a caption with model summary when meta-analysis is required.gghistostats
and itsgrouped_
variant has a new argumentnormal.curve
to superpose a normal distribution curve on top of the histogram (#138).ggcoefstats
can support following new regression model objects:brmsfit
,gam
,Gam
,gamlss
,mcmc
,mjoint
,stanreg
.New function to convert plots which are not of
gg
/ggplot
class toggplot
class objects.Instead of using
effsize
to compute Cohen’s d and Hedge’s g, ggstatsplot now relies on a new (#159) internal functioneffect_t_parametric
to compute them. This removeseffsize
from dependencies.To be consistent with other functions in this package, both
ggbarstats
andggpiestats
gainresults.subtitle
which can be set toFALSE
if statistical analysis is not required, in which casesubtitle
argument can be used to provide alternative subtitle.
MAJOR CHANGES
-
ggbetweenstats
now defaults to using noncentral-t distribution for computing Cohen’s d and Hedge’s g. To get variants with central-t distribution, useeffsize.noncentral = FALSE
.
MINOR CHANGES
All
grouped_
functions had argumenttitle.prefix
that defaulted to"Group"
. It now instead defaults toNULL
, in which case the prefix will variable name forgrouping.var
argument.To accommodate non-parametric tests,
subtitle_template
function can now work withparameter = NULL
.For
ggbetweenstats
, details contained in the subtitle for non-parametric test are modified. It now uses Spearman’s rho-based effect size estimates. This removescoin
from dependencies.ggbetweenstats
and itsgrouped_
variant gain a new argumentaxes.range.restrict
(which defaults toFALSE
). This restrictsy
-axes limits to minimum and maximum ofy
variable. This is what these functions were doing by default in the past versions, which created issues for additional ggplot components using theggplot.component
argument.All bayes factor related subtitle and captions replace
prior.width
withr_{Cauchy}
.ggcoefstats
passes dots (...
) toaugment
method frombroom
.
BUG FIXES
The helper function
bf_extractor
no longer provides option to extract information about posterior distribution because these details were incorrect. Theposterior = TRUE
details were not used anywhere in the package so nothing about the results changes.ggcorrmat
didn’t output pair names whenoutput == "ci"
was used. This is fixed.
ggstatsplot 0.0.8
CRAN release: 2019-01-07
NEW FEATURES
ggcoefstats
gainsmeta.analytic.effect
that can be used to carry out meta-analysis on regression estimates. This especially useful when a data frame with regression estimates and standard error is available from prior analyses. Thesubtitle
is prepared with the new functionsubtitle_meta_ggcoefstats
which is also exported.ggbetweenstats
,ggscatterstats
,gghistostats
, andggdotplotstats
(and theirgrouped_
variants) all gain a newggplot.component
argument. This argument will primarily be helpful to change the individual plots in agrouped_
plot.ggcoefstats
can support following new regression model objects:polr
,survreg
,cch
,Arima
,biglm
,glmmTMB
,coxph
,ridgelm
,aareg
,plm
,nlrq
,ivreg
,ergm
,btergm
,garch
,gmm
,lmodel2
,svyolr
,confusionMatrix
,multinom
,nlmerMod
,svyglm
,MCMCglmm
,lm.beta
,speedlm
,fitdistr
,mle2
,orcutt
,glmmadmb
.
BUG FIXES
ggcoefstats
didn’t work whenstatistic
argument was set toNULL
. This was not expected behavior. This has been fixed. Now, ifstatistic
is not specified, only the dot-and-whiskers will be shown without any labels.subtitle_t_parametric
was producing incorrect sample size information whenpaired = TRUE
and the data containedNA
s. This has been fixed.
MAJOR CHANGES
ggscatterstats
and itsgrouped_
variant accept both character and bare exressions as input to argumentslabel.var
andlabe.expression
(#110).To be consistent with rest of the functions in the package, both Pearson’s r, Spearman’s rho, and robust percentage bend correlations also display information about statistic associated with these tests.
ggscatterstats
, by default, showed jittered data points (because it relied onposition_jitter
defaults). This could be visually inaccurate and, therefore,ggscatterstats
now displays points without any jitter. The user can introduce jitter if they wish to usingpoint.width.jitter
andpoint.height.jitter
arguments. For similar reasons, forggbetweenstats
and itsgrouped_
variant,point.jitter.height
default has been changed from0.1
to0
(no vertical jitter, i.e.).
MINOR CHANGES
Confidence interval for Kendall’s W is now computed using
stats::kruskal.test
. As a result,PMCMRplus
removed from dependencies.ggcoefstats
gains acaption
argument. Ifcaption.summary
is set toTRUE
, the specified caption will be added on top of thecaption.summary
.
ggstatsplot 0.0.7
CRAN release: 2018-12-08
BUG FIXES
ggcoefstats
was showing wrong confidence intervals formerMod
class objects due to a bug in thebroom.mixed
package (https://github.com/bbolker/broom.mixed/issues/30#issuecomment-428385005). This was fixed inbroom.mixed
and soggcoefstats
should no longer have any issues.specify_decimal_p
has been modified because it produced incorrect results whenk < 3
andp.value = TRUE
(e.g.,0.002
was printed as< 0.001
).ggpiestats
produced incorrect results if some levels of the factor had been filtered out prior to using this function. It now drops unused levels and produces correct results.gghistostats
wasn’t filtering outNA
s properly. This has been fixed.
MAJOR CHANGES
New function
ggdotplotstats
for creating a dot plot/chart for labelled numeric data.All primary functions gain
conf.level
argument to control confidence level for effect size measures.As per APA guidelines, all results show results with two decimal places. That is, the default value for
k
argument for all functions has been changed from3
to2
.All helper functions for the
ggbetweenstats
subtitles have been renamed to remove_ggbetween_
from their names as this was becoming confusing for the user. Some of these functions work both with the between- and within-subjects designs, so having_ggbetween_
in their names made users suspect if they could use these functions for within-subjects designs.ggstatsplot now depends on
R 3.5.0
. This is because some of its dependencies require 3.5.0 to work (e.g.,broom.mixed
).All
theme_
functions are now exported (theme_pie()
,theme_corrmat()
).ggbetweenstats
now supports multiple pairwise comparison tests (parametric, nonparametric, and robust variants). It gains a new dependencyggsignif
.ggbetweenstats
now supports eta-squared and omega-squared effect sizes for anova models. This function gains a new argumentpartial
.Following functions are now reexported from the
groupedstats
package to avoid repeating the same code in two packages:specify_decimal_p
,signif_column
,lm_effsize_ci
, andset_cwd
. Therefore,groupedstats
is now added as a dependency.gghistostats
can now show both counts and proportions information on the same plot whenbar.measure
argument is set to"mix"
.ggcoefstats
works with tidy dataframes.The helper function
untable
has been deprecated in light oftidyr::uncount
, which does exactly whatuntable
was doing. The author wasn’t aware of this function whenuntable
was written.All vignettes have been removed from
CRAN
to reduce the size of the package. They are now available on the package website: https://indrajeetpatil.github.io/ggstatsplot/articles/.subtitle_t_robust
function can now handle dependent samples and gainspaired
argument.A number of tidyverse operators are now reexported by ggstatsplot:
%>%
,%<>%
,%$%
.
MINOR CHANGES
ggscatterstats
,ggpiestats
, and theirgrouped_
variant support bayes factor tests and gain new arguments relevant to this test.Effect size and their confidence intervals now available for Kruskal-Wallis test.
Minor stylistic changes to how symbols for partial-eta-/omega-squared were being displayed in subtitles.
ggbetweenstats
supports bayes factor tests for anova designs.ggpiestats
(and itsgrouped_
version) gainslice.label
argument that decides what information needs to be displayed as a label on the slices of the pie chart:"percentage"
(which has been the default thus far),"counts"
, or"both"
.ggcorrmat
can work withcor.vars = NULL
. In such case, all numeric variables from the provided data frame will be used for computing the correlation matrix.Given the constant changes to the default behavior of functions, the lifecycle badge has been changed from
stable
tomaturing
.When the number of colors needed by a function exceeds the number of colors contained in a given palette, informative message is displayed to the user (with the new internal function
palette_message()
).Several users had requested an easier way to turn off subtitles with results from tests (which was already implemented in
ggscatterstats
andgghistostats
with the argumentresults.subtitle
), soggbetweenstats
also gains two new arguments to do this:results.subtitle
andsubtitle
.New dataset added:
iris_long
.More tests added and the code coverage has now jumped to over 75%.
To avoid code repetition, there is a now a function that produces a generic message any time confidence intervals for effect size estimate are computed using bootstrapping.
ggstatsplot 0.0.6
CRAN release: 2018-09-30
MAJOR CHANGES
The package now exports all functions used to create text expressions with results. This makes it easy for people to use these results in their own plots at any location they want (and not just in
subtitle
, the current default for ggstatsplot).ggcorrmat
gainsp.adjust.method
argument which allows p-values for correlations to be corrected for multiple comparisons.ggscatterstats
gainslabel.var
andlabel.expression
arguments to attach labels to points.gghistostats
now defaults to not showing (redundant) color gradient (fill.gradient = FALSE
) and shows both"count"
and"proportion"
data. It also gains a new argumentbar.fill
that can be used to fill bars with a uniform color.ggbetweenstats
,ggcoefstats
,ggcorrmat
,ggscatterstats
, andggpiestats
now support all palettes contained in thepaletteer
package. This helps avoid situations where people had large number of groups (> 12) and there were not enough colors in any of theRColorBrewer
palettes.ggbetweenstats
gainsbf.message
argument to display bayes factors in favor of the null (currently works only for parametric t-test).gghistostats
function no longer hasline.labeller.y
argument; this position is automatically determined now.
BREAKING CHANGES
legend.title.margin
function has been deprecated sinceggplot2 3.0.0
has improved on the margin issues from previous versions. All functions that wrapped around this function now lose the relevant arguments (legend.title.margin
,t.margin
,b.margin
).The argument
ggstatsplot.theme
has been changed toggstatsplot.layer
forggcorrmat
function to be consistent across functions.For consistency,
conf.level
andconf.type
arguments forggbetweenstats
have been deprecated. No other function in the package allowed changing confidence interval or their type for effect size estimation. These arguments were relevant only forrobust
tests anyway.ggocorrmat
argumenttype
has been changed tomatrix.type
because for all other functionstype
argument specifies the type of the test, while for this function it specified the display of the visualization matrix. This will make the syntax more consistent across functions.ggscatterstats
gains new arguments to specify aesthetics for geom point (point.color
,point.size
,point.alpha
). To be consistent with this naming schema, thewidth.jitter
andheight.jitter
arguments have been renamed topoint.width.jitter
andpoint.height.jitter
, resp.
MINOR CHANGES
gghistostats
: To be compatible withJASP
, natural logarithm of Bayes Factors is displayed, and not base 10 logarithm.ggscatterstats
gainsmethod
andformula
arguments to modify smoothing functions.ggcorrmat
can now showrobust
correlation coefficients in the matrix plot.For
gghistostats
,binwidth
value, if not specified, is computed with(max-min)/sqrt(n)
. This is basically to get rid of the warnings ggplot2 produces. Thanks to Chuck Powell’s PR (#43).ggcoefstats
gains a new argumentpartial
and can display eta-squared and omega-squared effect sizes for anovas, in addition to the prior partial variants of these effect sizes.ggpiestats
gainsdigits.perc
argument to show desired number of decimal places in percentage labels.
ggstatsplot 0.0.5
CRAN release: 2018-08-14
MAJOR CHANGES
For the sake of consistency,
theme_mprl
is now calledtheme_ggstatsplot
. Thetheme_mprl
function will still be around and will not be deprecated, so feel free to use either or both of them since they are identical.ggcoefstats
no longer has argumentseffects
andran_params
because only fixed effects are shown for mixed-effects models.ggpiestats
can now handle within-subjects designs (McNemar test results will be displayed).
BUG FIXES
ggbetweenstats
was producing wrong axes labels whensample.size.label
was set toTRUE
and user had reordered factor levels before using this function. The new version fixes this.ggcoefstats
wasn’t producing partial omega-squared foraovlist
objects. Fixed that with new version ofsjstats
.
MINOR CHANGES
Removed the trailing comma from the robust correlation analyses.
gghistostats
has a new argument to remove color fill gradient.ggbetweenstats
takes new argumentmean.ci
to show confidence intervals for the mean values.For
lmer
models, p-values are now computed usingsjstats::p_value
. This removeslmerTest
package from dependencies.sjstats
no longer suggestsapaTables
package to compute confidence intervals for partial eta- and omega-squared. Therefore,apaTables
andMBESS
are removed from dependencies.ggscatterstats
supportsdensigram
with the development version ofggExtra
. It additionally gains few extra arguments to change aesthetics of marginals (alpha, size, etc.).
ggstatsplot 0.0.4
CRAN release: 2018-07-05
MAJOR CHANGES
New function:
ggcoefstats
for displaying model coefficients.All functions now have
ggtheme
argument that can be used to change the default theme, which has now been changed fromtheme_grey()
totheme_bw()
.The robust correlation is no longer
MASS::rlm
, but percentage bend correlation, as implemented inWRS2::pbcor
. This was done to be consistent across different functions.ggcorrmat
also uses percentage bend correlation as the robust correlation measure. This also means that ggstatsplot no longer importsMASS
andsfsmisc
.The
data
argument is no longerNULL
for all functions, exceptgghistostats
. In other words, the user must provide a data frame from which variables or formulas should be selected.All subtitles containing results now also show sample size information (n). To adjust for the inflated length of the subtitle, the default subtitle text size has been changed from
12
to11
.
MINOR CHANGES
Switched back to Shapiro-Wilk test of normality to remove
nortest
from imports.ggbetweenstats
andggpiestats
now display sample sizes for each level of the groping factor by default. This behavior can be turned off by settingsample.size.label
toFALSE
.Three new datasets added:
Titanic_full
,movies_wide
,movies_long
.Added confidence interval for effect size for robust ANOVA.
The 95% CI for Cramer’V computed using
boot::boot
. Therefore, the package no longer importsDescTools
.To be consistent across correlations covered, all correlations now show estimates for correlation coefficients, confidence intervals for the estimate, and p-values. Therefore, t-values and regression coefficients are no longer displayed for Pearson’s r.
The
legend.title.margin
arguments forgghistostats
andggcorrmat
now default toFALSE
, sinceggplot2 3.0.0
has better legend title margins.ggpiestats
now sorts the summary dataframes not by percentages but by the levels ofmain
variable. This was done to have the same legends across different levels of a grouping variable ingrouped_ggpiestats
.To remove cluttered display of results in the subtitle,
ggpiestats
no longer shows titles for the tests run (these were “Proportion test” and “Chi-Square test”). From the pie charts, it should be obvious to the user or reader what test was run.gghistostats
also allows running robust version of one-sample test now (One-sample percentile bootstrap).
ggstatsplot 0.0.3
CRAN release: 2018-05-22
NEW FEATURES
The
ggbetweenstats
function can now show notched box plots. Two new argumentsnotch
andnotchwidth
control its behavior. The defaults are still standard box plots.Removed warnings that were appearing when
outlier.label
argument was ofcharacter
type.The default color palette used for all plots is colorblind friendly.
gghistostats
supportsproportion
anddensity
as a value measure for bar heights to show proportions and density. New argumentbar.measure
controls this behavior.grouped_
variants of functionsggcorrmat
,ggscatterstats
,ggbetweenstats
, andggpiestats
introduced to create multiple plots for different levels of a grouping variable.
MAJOR CHANGES
To be internally consistent, all functions in ggstatsplot use the spelling
color
, rather thancolour
in some functions, whilecolor
in others.Removed the redundant argument
binwidth.adjust
fromgghistostats
function. This argument was relevant for the first avatar of this function, but is no longer playing any role.To be internally consistent, the argument
lab_col
andlab_size
inggcorrmat
have been changed tolab.col
andlab.size
, respectively.
MINOR CHANGES
Added a new argument to
ggstatsplot.theme
function to control ifggstatsplot::theme_mprl
is to be overlaid on top of the selectedggtheme
(ggplot2 theme, i.e.).Two new arguments added to
gghistostats
to allow user to change colorbar gradient. Defaults are colorblind friendly.Both
gghistostats
andggcorrmat
have a new argumentlegend.title.margin
to control margin adjustment between the title and the colorbar.The vertical lines denoting test values and centrality parameters can be tagged with text labels with a new argument
line.labeller
ingghistostats
function.
ggstatsplot 0.0.2
CRAN release: 2018-04-28
NEW FEATURES
Bayesian test added to
gghistostats
and two new arguments to also display a vertical line fortest.value
argument.Vignette added for
gghistostats
.Added new function
grouped_gghistostats
to facilitate applyinggghistostats
for multiple levels of a grouping factor.ggbetweenstats
has a new argumentoutlier.coef
to adjust threshold used to detect outliers. Removed bug from the same function whenoutlier.label
argument is of factor/character type.
MAJOR CHANGES
Functions
signif_column
andgrouped_proptest
are now deprecated. They were exported in the first release by mistake.Function
gghistostats
no longer displays both density and count since the density information was redundant. Thedensity.plot
argument has also been deprecated.ggscatterstats
argumentintercept
has now been changed tocentrality.para
. This was due to possible confusion about interpretation of these lines; they show central tendency measures and not intercept for the linear model. Thus the change.The default for
effsize.type = "biased"
effect size forggbetweenstats
in case of ANOVA is partial omega-squared, and not omega-squared. Additionally, both partial eta- and omega-squared are not computed using bootstrapping with (default) 100 bootstrap samples.
MINOR CHANGES
More examples added to the
README
document.95% confidence intervals for Spearman’s rho are now computed using
broom
package.RVAideMemoire
package is thus removed from dependencies.95% confidence intervals for partial eta- and omega-squared for
ggbetweenstats
function are now computed usingsjstats
package, which allows bootstrapping.apaTables
anduserfriendlyscience
packages are thus removed from dependencies.