Merge pull request #384 from easystats/CRAN-0.5-for-real-this-time

Cran 0.5 for real this time

.gitignore

@@ -2,6 +2,9 @@
 .Rhistory
 .Rapp.history
 
+# revdepcheck
+/revdep*
+
 # Session Data files
 .RData
 

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: effectsize
 Title: Indices of Effect Size and Standardized Parameters
-Version: 0.4.99
+Version: 0.5
 Authors@R: 
     c(person(given = "Mattan S.",
              family = "Ben-Shachar",
@@ -54,7 +54,7 @@ Imports:
     bayestestR (>= 0.10.5),
     insight (>= 0.14.3),
     parameters (>= 0.14.0),
-    performance (>= 0.7.3.5),
+    performance (>= 0.8.0),
     datawizard (>= 0.2.0),
     stats,
     utils
@@ -91,6 +91,5 @@ Encoding: UTF-8
 Language: en-US
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.1.2
-Remotes: easystats/performance
 Config/testthat/edition: 3
 Config/testthat/parallel: true

NEWS.md

@@ -6,7 +6,7 @@
 - `interpret_d()`, `interpret_g()`, and `interpret_delta()` are now `interpret_cohens_d()`, `interpret_hedges_g()`, and `interpret_glass_delta()`.
 - `interpret_parameters()` was removed. Use `interpret_r()` instead (with caution!).
 - Phi, Cohen's *w*, Cramer's *V*, ANOVA effect sizes, rank Epsilon squared, Kendall's *W* - CIs default to 95% one-sided CIs (`alternative = "greater"`). (To restore previous behavior, set `ci = .9, alternative = "two.sided"`.)
-- `adjust()`, `change_scale()`, `normalize()`, `ranktransform()`, `standardize()` (data), and `unstandardize()` have moved to the new [`{datawizard}`](https://easystats.github.io/datawizard) package!
+- `adjust()`, `change_scale()`, `normalize()`, `ranktransform()`, `standardize()` (data), and `unstandardize()` have moved to the new [`{datawizard}`](https://easystats.github.io/datawizard/) package!
 
 ## New features
 

R/docs_extra.R

@@ -5,17 +5,14 @@
 #'
 #' @section Confidence (Compatibility) Intervals (CIs):
 #' Unless stated otherwise, confidence (compatibility) intervals (CIs) are
-#' estimated using the noncentrality parameter method (also called the
-#' "pivot method"). This method finds the noncentrality parameter ("*ncp*") of
-#' a noncentral *t*, *F*, or
-#' \ifelse{latex}{\out{$\chi^2$}}{\ifelse{html}{\out{&chi;<sup>2</sup>}}{chi-squared}}
-#' distribution that places the observed *t*, *F*, or
-#' \ifelse{latex}{\out{$\chi^2$}}{\ifelse{html}{\out{&chi;<sup>2</sup>}}{chi-squared}}
-#' test statistic at the desired probability point of the distribution.
-#' For example, if the observed *t* statistic is 2.0, with 50
-#' degrees of freedom, for which cumulative noncentral *t* distribution is
-#' *t* = 2.0 the .025 quantile (answer: the noncentral *t* distribution with
-#' *ncp* = .04)? After estimating these confidence bounds on the *ncp*, they are
+#' estimated using the noncentrality parameter method (also called the "pivot
+#' method"). This method finds the noncentrality parameter ("*ncp*") of a
+#' noncentral *t*, *F*, or \eqn{\chi^2} distribution that places the observed
+#' *t*, *F*, or \eqn{\chi^2} test statistic at the desired probability point of
+#' the distribution. For example, if the observed *t* statistic is 2.0, with 50
+#' degrees of freedom, for which cumulative noncentral *t* distribution is *t* =
+#' 2.0 the .025 quantile (answer: the noncentral *t* distribution with *ncp* =
+#' .04)? After estimating these confidence bounds on the *ncp*, they are
 #' converted into the effect size metric to obtain a confidence interval for the
 #' effect size (Steiger, 2004).
 #' \cr\cr
@@ -26,23 +23,20 @@
 #' in a hypothesis test, and more." (Steiger, 2004). Confidence (compatibility)
 #' intervals and p values are complementary summaries of parameter uncertainty
 #' given the observed data. A dichotomous hypothesis test could be performed
-#' with either a CI or a p value. The
-#' 100(\ifelse{latex}{\out{$1 - \alpha$}}{\ifelse{html}{\out{1 &minus; &alpha;}}{1 - alpha}})%
-#' confidence interval contains all of the parameter values for which
-#' \ifelse{latex}{\out{$p > \alpha$}}{\ifelse{html}{\out{p > &alpha;}}{p > alpha}}
+#' with either a CI or a p value. The 100 (1 - \eqn{\alpha})% confidence
+#' interval contains all of the parameter values for which *p* > \eqn{\alpha}
 #' for the current data and model. For example, a 95% confidence interval
 #' contains all of the values for which p > .05.
 #' \cr\cr
 #' Note that a confidence interval including 0 *does not* indicate that the null
 #' (no effect) is true. Rather, it suggests that the observed data together with
 #' the model and its assumptions combined do not provided clear evidence against
 #' a parameter value of 0 (same as with any other value in the interval), with
-#' the level of this evidence defined by the chosen
-#' \ifelse{latex}{\out{$\alpha$}}{\ifelse{html}{\out{&alpha;}}{alpha}} level
-#' (Rafi & Greenland, 2020; Schweder & Hjort, 2016; Xie & Singh, 2013). To infer
-#' no effect, additional judgments about what parameter values are "close
-#' enough" to 0 to be negligible are needed ("equivalence testing"; Bauer &
-#' Kiesser, 1996).
+#' the level of this evidence defined by the chosen \eqn{\alpha} level (Rafi &
+#' Greenland, 2020; Schweder & Hjort, 2016; Xie & Singh, 2013). To infer no
+#' effect, additional judgments about what parameter values are "close enough"
+#' to 0 to be negligible are needed ("equivalence testing"; Bauer & Kiesser,
+#' 1996).
 #'
 #' @section One-Sided CIs:
 #' Typically, CIs are constructed as two-tailed intervals, with an equal
@@ -59,51 +53,39 @@
 #' are generally tested using 2-tailed tests and 2-sided CIs.
 #' \cr\cr
 #' Some effect sizes are strictly positive--they do have a minimum value, of 0.
-#' For example,
-#' \ifelse{latex}{\out{$R^2$}}{\ifelse{html}{\out{<i>R</i><sup>2</sup>}}{R^2}},
-#' \ifelse{latex}{\out{$\eta^2$}}{\ifelse{html}{\out{&eta;<sup>2</sup>}}{eta^2}},
-#' and other variance-accounted-for effect sizes, as well as Cramer's *V* and
-#' multiple *R*, range from 0 to 1. These typically involve *F*- or
-#' \ifelse{latex}{\out{$\chi^2$}}{\ifelse{html}{\out{&chi;<sup>2</sup>}}{chi-squared}}-statistics
-#' and are generally tested using *1-tailed* tests which test whether the
-#' estimated effect size is *larger* than the hypothesized null value (e.g., 0).
-#' In order for a CI to yield the same significance decision it must then by a
-#' *1-sided* CI, estimating only a lower bound. This is the default CI computed
-#' by *effectsize* for these effect sizes, where `alternative = "greater"` is
-#' set.
+#' For example, \eqn{R^2}, \eqn{\eta^2}, and other variance-accounted-for effect
+#' sizes, as well as Cramer's *V* and multiple *R*, range from 0 to 1. These
+#' typically involve *F*- or \eqn{\chi^2}-statistics and are generally tested
+#' using *1-tailed* tests which test whether the estimated effect size is
+#' *larger* than the hypothesized null value (e.g., 0). In order for a CI to
+#' yield the same significance decision it must then by a *1-sided* CI,
+#' estimating only a lower bound. This is the default CI computed by
+#' *effectsize* for these effect sizes, where `alternative = "greater"` is set.
 #' \cr\cr
 #' This lower bound interval indicates the smallest effect size that is not
 #' significantly different from the observed effect size. That is, it is the
 #' minimum effect size compatible with the observed data, background model
-#' assumptions, and
-#' \ifelse{latex}{\out{$\alpha$}}{\ifelse{html}{\out{&alpha;}}{alpha}} level.
-#' This type of interval does not indicate a maximum effect size value; anything
-#' up to the maximum possible value of the effect size (e.g., 1) is in the
-#' interval.
+#' assumptions, and \eqn{\alpha} level. This type of interval does not indicate
+#' a maximum effect size value; anything up to the maximum possible value of the
+#' effect size (e.g., 1) is in the interval.
 #' \cr\cr
 #' One-sided CIs can also be used to test against a maximum effect size value
-#' (e.g., is
-#' \ifelse{latex}{\out{$R^2$}}{\ifelse{html}{\out{<i>R</i><sup>2</sup>}}{R^2}}
-#' significantly smaller than a perfect correlation of 1.0?) can by setting
-#' `alternative = "less"`. This estimates a CI with only an *upper* bound;
-#' anything from the minimum possible value of the effect size (e.g., 0) up to
-#' this upper bound is in the interval.
+#' (e.g., is \eqn{R^2} significantly smaller than a perfect correlation of 1.0?)
+#' can by setting `alternative = "less"`. This estimates a CI with only an
+#' *upper* bound; anything from the minimum possible value of the effect size
+#' (e.g., 0) up to this upper bound is in the interval.
 #' \cr\cr
 #' We can also obtain a 2-sided interval by setting `alternative = "two-sided"`.
 #' These intervals can be interpreted in the same way as other 2-sided
 #' intervals, such as those for *r*, *d*, or *g*.
 #' \cr\cr
 #' An alternative approach to aligning significance tests using CIs and 1-tailed
 #' *p* values that can often be found in the literature is to construct a
-#' 2-sided CI at a lower confidence level (e.g., 100(\ifelse{latex}{\out{$1 -
-#' 2\alpha$}}{\ifelse{html}{\out{1 &minus; 2&alpha;}}{1 - 2*alpha}})% =
-#' \ifelse{latex}{\out{$100 - 2 \times 5\% = 90\%$}}{\ifelse{html}{\out{100
-#' &minus; 2 &times; 5&percnt; = 90&percnt;}}{100 - 2*5% = 90%}}). This
-#' estimates the lower bound and upper bound for the above 1-sided intervals
-#' simultaneously. These intervals are commonly reported when conducting
-#' **equivalence tests**. For example, a 90% 2-sided interval gives the bounds
-#' for an equivalence test with \ifelse{latex}{\out{$\alpha =
-#' .05$}}{\ifelse{html}{\out{&alpha; = .05}}{alpha = .05}}. However, be aware
+#' 2-sided CI at a lower confidence level (e.g., 100(1-2\eqn{\alpha})% = 100 -
+#' 2*5% = 90%. This estimates the lower bound and upper bound for the above
+#' 1-sided intervals simultaneously. These intervals are commonly reported when
+#' conducting **equivalence tests**. For example, a 90% 2-sided interval gives
+#' the bounds for an equivalence test with \eqn{\alpha} = .05. However, be aware
 #' that this interval does not give 95% coverage for the underlying effect size
 #' parameter value. For that, construct a 95% 2-sided CI.
 #'

R/standardize.models.R

@@ -3,6 +3,14 @@
 #' Performs a standardization of data (z-scoring) using
 #' [`datawizard::standardize()`] and then re-fits the model to the standardized
 #' data.
+#' \cr\cr
+#' Standardization is done by completely refitting the model on the standardized
+#' data. Hence, this approach is equal to standardizing the variables *before*
+#' fitting the model and will return a new model object. This method is
+#' particularly recommended for complex models that include interactions or
+#' transformations (e.g., polynomial or spline terms). The `robust` (default to
+#' `FALSE`) argument enables a robust standardization of data, based on the
+#' `median` and the `MAD` instead of the `mean` and the `SD`.
 #'
 #' @param x A statistical model.
 #' @param weights If `TRUE` (default), a weighted-standardization is carried out.
@@ -16,11 +24,37 @@
 #'
 #' @return A statistical model fitted on standardized data
 #'
-#' @inheritSection standardize_parameters Generalized Linear Models
+#' @details
 #'
+#' # Generalized Linear Models
+#' Standardization for generalized linear models (GLM, GLMM, etc) is done only
+#' with respect to the predictors (while the outcome remains as-is,
+#' unstandardized) - maintaining the interpretability of the coefficients (e.g.,
+#' in a binomial model: the exponent of the standardized parameter is the OR of
+#' a change of 1 SD in the predictor, etc.)
+#'
+#' # Dealing with Factors
+#' `standardize(model)` or `standardize_parameters(model, method = "refit")` do
+#' *not* standardized categorical predictors (i.e. factors) / their
+#' dummy-variables, which may be a different behaviour compared to other R
+#' packages (such as \pkg{lm.beta}) or other software packages (like SPSS). To
+#' mimic such behaviours, either use `standardize_parameters(model, method =
+#' "basic")` to obtain post-hoc standardized parameters, or standardize the data
+#' with `datawizard::standardize(data, force = TRUE)` *before* fitting the
+#' model.
+#'
+#' # Transformed Variables
+#' When the model's formula contains transformations (e.g. `y ~ exp(X)`) the
+#' transformation effectively takes place after standardization (e.g.,
+#' `exp(scale(X))`). Since some transformations are undefined for none positive
+#' values, such as `log()` and `sqrt()`, the releven variables are shifted (post
+#' standardization) by `Z - min(Z) + 1` or `Z - min(Z)` (respectively).
+#'
+#' @family standardize
 #' @examples
 #' model <- lm(Infant.Mortality ~ Education * Fertility, data = swiss)
 #' coef(standardize(model))
+#'
 #' @importFrom stats update
 #' @importFrom insight get_data model_info find_response get_response find_weights get_weights
 #' @importFrom datawizard standardize

R/standardize_parameters.R

@@ -74,33 +74,25 @@
 #' level 1 predictors (Hoffman 2015, page 342). A warning is given when a
 #' within-group varialbe is found to have access between-group variance.
 #'
-#' ## Dealing with Factors
-#' The `"refit"` method does *not* standardized categorical predictors (i.e.
-#' factors), which may be a different behaviour compared to other R packages
-#' (such as \pkg{lm.beta}) or other software packages (like SPSS). to mimic such
-#' behaviours, either use the `"basic"` method or standardize the data with
-#' `effectsize::standardize(force=TRUE)` *before* fitting the model.
-#'
-#' ## Transformed Variables
-#' When the model's formula contains transformations (e.g. `y ~ exp(X)`)
-#' `method = "refit"` might give different results compared to
-#' `method = "basic"` (`"posthoc"` and `"smart"` do not support such
-#' transformations): where `"refit"` standardizes the data prior to the
-#' transformation (e.g. equivalent to `exp(scale(X))`), the `"basic"` method
-#' standardizes the transformed data (e.g. equivalent to `scale(exp(X))`). See
-#' [standardize()] for more details on how different transformations are dealt
-#' with.
+#' # Transformed Variables
+#' When the model's formula contains transformations (e.g. `y ~ exp(X)`) `method
+#' = "refit"` will give different results compared to `method = "basic"`
+#' (`"posthoc"` and `"smart"` do not support such transformations): While
+#' `"refit"` standardizes the data *prior* to the transformation (e.g.
+#' equivalent to `exp(scale(X))`), the `"basic"` method standardizes the
+#' transformed data (e.g. equivalent to `scale(exp(X))`).
+#' \cr\cr
+#' See the *Transformed Variables* section in [standardize.default()] for more
+#' details on how different transformations are dealt with when `method =
+#' "refit"`.
 #'
 #' # Confidence Intervals
 #' The returned confidence intervals are re-scaled versions of the
 #' unstandardized confidence intervals, and not "true" confidence intervals of
 #' the standardized coefficients (cf. Jones & Waller, 2015).
 #'
-#' # Generalized Linear Models
-#' When standardizing coefficients of a generalized model (GLM, GLMM, etc), only
-#' the predictors are standardized, maintaining the interpretability of the
-#' coefficients (e.g., in a binomial model: the exponent of the standardized
-#' parameter is the OR of a change of 1 SD in the predictor, etc.)
+#' @inheritSection standardize.default Generalized Linear Models
+#' @inheritSection standardize.default Dealing with Factors
 #'
 #' @return A data frame with the standardized parameters (`Std_*`, depending on
 #'   the model type) and their CIs (`CI_low` and `CI_high`). Where applicable,
@@ -109,7 +101,6 @@
 #'
 #' @family standardize
 #' @family effect size indices
-#' @seealso [standardize_info()]
 #'
 #' @examples
 #' library(effectsize)

cran-comments.md

@@ -4,7 +4,7 @@
 * GitHub Actions (windows):      devel, release, oldrel
 * Github Actions (macOS):        devel, release, oldrel
 * GitHub Actions (ubuntu-16.04): devel, release, oldrel, 3.6, 3.5, 3.4
-* win-builder:                   devel
+* win-builder:                   release
 
 
 ## R CMD check results
@@ -17,6 +17,9 @@
 
 
 
-## Reverse Dependancies 
+## `revdepcheck` results
 
-- The maintainer of `statsExperssions` has been informed about the imminent failure of tests with this new version.
+We checked 15 reverse dependencies, comparing R CMD check results across CRAN and dev versions of this package.
+
+ * We saw 0 new problems
+ * We failed to check 0 packages