Cran 0.7.0.5 (#465)

## Breaking Changes

`effectsize` now requires minimal *`R`* version of `3.5`.

## Bug fixes

- `cohens_d()` for paired / one sample now gives more accurate CIs (was off by a factor of `(N - 1) / N`; #457)
- `kendalls_w()` now deals correctly with singular ties (#448).

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: effectsize
 Title: Indices of Effect Size and Standardized Parameters
-Version: 0.7.0.1
+Version: 0.7.0.5
 Authors@R: 
     c(person(given = "Mattan S.",
              family = "Ben-Shachar",
@@ -59,16 +59,10 @@ Imports:
     bayestestR (>= 0.12.1),
     insight (>= 0.18.0),
     parameters (>= 0.18.1),
-    performance (>= 0.9.0),
-    datawizard (>= 0.4.1.10),
+    performance (>= 0.9.1),
+    datawizard (>= 0.5.0),
     stats,
     utils
-Remotes: 
-    easystats/datawizard, 
-    easystats/parameters, 
-    easystats/performance,
-    easystats/correlation,
-    easystats/bayestestR
 Suggests:
     correlation (>= 0.8.1),
     see (>= 0.7.1),

NEWS.md

@@ -1,17 +1,12 @@
-# effectsize 0.7.1.9990
-
-## Bug fixes
-
-- `cohens_d()` for paired / one sample now gives more accurate CIs (was off by a factor of `(N - 1) / N`; #457)
-
-# effectsize 0.7.0.9
+# effectsize 0.7.0.5
 
 ## Breaking Changes
 
 `effectsize` now requires minimal *`R`* version of `3.5`.
 
 ## Bug fixes
 
+- `cohens_d()` for paired / one sample now gives more accurate CIs (was off by a factor of `(N - 1) / N`; #457)
 - `kendalls_w()` now deals correctly with singular ties (#448).  
 
 # effectsize 0.7.0

R/effectsize.htest.R

@@ -54,7 +54,8 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
     args <- list(
       t = unname(model$statistic),
       df_error = unname(model$parameter),
-      paired = !grepl("Two", model$method)
+      paired = !grepl("Two", model$method),
+      verbose = verbose
     )
   } else {
     if (grepl(" by ", model$data.name, fixed = TRUE)) {
@@ -144,6 +145,7 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
       nrow = nr,
       ncol = nc,
       p = p,
+      verbose = verbose,
       ...
     )
   } else {
@@ -198,6 +200,7 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
     f = model$statistic,
     df = model$parameter[1],
     df_error = model$parameter[2],
+    verbose = verbose,
     ...
   )
   colnames(out)[1] <- sub("_partial", "", colnames(out)[1])
@@ -216,9 +219,9 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
   .fail_if_approx(approx, "cohens_g")
 
   if (inherits(data, "table")) {
-    out <- cohens_g(data, ...)
+    out <- cohens_g(data, verbose = verbose, ...)
   } else {
-    out <- cohens_g(data[[1]], data[[2]], ...)
+    out <- cohens_g(data[[1]], data[[2]], verbose = verbose, ...)
   }
   out
 }
@@ -251,7 +254,8 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
   args <- list(
     x = data[[1]],
     y = if (ncol(data) == 2) data[[2]],
-    paired = grepl("signed rank", model$method, fixed = TRUE)
+    paired = grepl("signed rank", model$method, fixed = TRUE),
+    verbose = verbose
   )
 
   if (type == "cles") {
@@ -278,10 +282,10 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
 
 
   if (inherits(data, "data.frame")) {
-    out <- rank_epsilon_squared(data[[1]], data[[2]], ...)
+    out <- rank_epsilon_squared(data[[1]], data[[2]], verbose = verbose, ...)
   } else {
     # data frame
-    out <- rank_epsilon_squared(data, ...)
+    out <- rank_epsilon_squared(data, verbose = verbose, ...)
   }
   out
 }
@@ -300,7 +304,7 @@ effectsize.htest <- function(model, type = NULL, verbose = TRUE, ...) {
     data <- as.data.frame(data)[c("Freq", "Var2", "Var1")]
   }
 
-  out <- kendalls_w(data[[1]], data[[2]], data[[3]], ...)
+  out <- kendalls_w(data[[1]], data[[2]], data[[3]], verbose = verbose, ...)
   out
 }
 

R/eta_squared.R

@@ -111,8 +111,6 @@
 #' epsilon_squared(model)
 #' cohens_f(model)
 #'
-#' if (require(see)) plot(eta2)
-#'
 #' model0 <- aov(mpg ~ am_f + cyl_f, data = mtcars) # no interaction
 #' cohens_f_squared(model0, model2 = model)
 #'
@@ -125,7 +123,10 @@
 #'
 #' interpret(eta2, rules = "cohen1992")
 #'
-#' @examplesIf require("car") && require("afex")
+#' @examplesIf require("see") && FALSE
+#' plot(eta2) # Requires the {see} package
+#'
+#' @examplesIf require("car")
 #' # Recommended: Type-2 or -3 effect sizes + effects coding
 #' # -------------------------------------------------------
 #' contrasts(mtcars$am_f) <- contr.sum
@@ -134,28 +135,27 @@
 #' model <- aov(mpg ~ am_f * cyl_f, data = mtcars)
 #' model_anova <- car::Anova(model, type = 3)
 #'
-#' eta_squared(model_anova)
+#' epsilon_squared(model_anova)
 #'
+#' @examplesIf require("car") && require("afex")
 #' # afex takes care of both type-3 effects and effects coding:
 #' data(obk.long, package = "afex")
 #' model <- afex::aov_car(value ~ treatment * gender + Error(id / (phase)),
 #'   data = obk.long, observed = "gender"
 #' )
-#' eta_squared(model)
-#' epsilon_squared(model)
+#'
 #' omega_squared(model)
 #' eta_squared(model, generalized = TRUE) # observed vars are pulled from the afex model.
 #'
-#' @examplesIf require("lmerTest") && require("lme4")
+#' @examplesIf require("lmerTest") && require("lme4") && FALSE
 #' ## Approx. effect sizes for mixed models
 #' ## -------------------------------------
 #' model <- lme4::lmer(mpg ~ am_f * cyl_f + (1 | vs), data = mtcars)
 #' omega_squared(model)
 #'
-#' @examplesIf require(rstanarm) && require(bayestestR) && require(car)
+#' @examplesIf require(rstanarm) && require(bayestestR) && require(car) && FALSE
 #' ## Bayesian Models (PPD)
 #' ## ---------------------
-#' \dontrun{
 #' fit_bayes <- rstanarm::stan_glm(
 #'   mpg ~ factor(cyl) * wt + qsec,
 #'   data = mtcars, family = gaussian(),
@@ -175,7 +175,6 @@
 #' )
 #' aov_table <- car::Anova(fit_freq, type = 3)
 #' eta_squared(aov_table)
-#' }
 #'
 #' @return A data frame containing the effect size values and their confidence
 #'   intervals.

R/interpret_bf.R

@@ -39,8 +39,7 @@
 #' Sociological methodology, 25, 111-164.
 #'
 #' - Jarosz, A. F., & Wiley, J. (2014). What are the odds? A practical guide to
-#' computing and reporting Bayes factors. The Journal of Problem Solving, 7(1),
-#' 2.
+#' computing and reporting Bayes factors. The Journal of Problem Solving, 7(1), 2.
 #'
 #' @export
 interpret_bf <- function(bf,

cran-comments.md

@@ -1,12 +1,13 @@
 All URL issues have been resolved.
+DOI issues are a false positive.
 
 ## Test environments
 
-* local installation: R 4.1.1 on Windows
+* local installation: R 4.2.1 on Windows
 * GitHub Actions
     - Windows:        devel, release, oldrel
-    - macOS:          devel, release, oldrel
-    - ubuntu-16.04:   devel, release, oldrel, 4.0, 3.6, 3.5
+    - macOS:          devel, release, oldrel, 4.0, 3.6, 3.5
+    - ubuntu-18.04:   devel, release, oldrel
 * win-builder:        release
 
 
@@ -23,7 +24,7 @@ All URL issues have been resolved.
 
 ## revdepcheck results
 
-We checked 17 reverse dependencies, comparing R CMD check results across CRAN and dev versions of this package.
+We checked 16 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

tests/testthat/test-effectsize.R

@@ -171,8 +171,9 @@ if (require("testthat") && require("effectsize")) {
       )
       )
     ft <- friedman.test(RoundingTimes)
-    expect_equal(effectsize(ft)[[1]], W <- kendalls_w(RoundingTimes)[[1]], ignore_attr = TRUE)
-    expect_equal(kendalls_w(ft)[[1]], W, ignore_attr = TRUE)
+    W <- kendalls_w(RoundingTimes, verbose = FALSE, ci = NULL)
+    expect_equal(effectsize(ft, verbose = FALSE, ci = NULL), W, ignore_attr = TRUE)
+    expect_equal(kendalls_w(ft, verbose = FALSE, ci = NULL), W, ignore_attr = TRUE)
 
     X <<- c(2.9, 3.0, 2.5, 2.6, 3.2) # normal subjects
     Y <<- c(3.8, 2.7, 4.0, 2.4) # with obstructive airway disease

tests/testthat/test-rankES.R

@@ -116,7 +116,7 @@ if (require("testthat") && require("effectsize")) {
     )
     dat
 
-    W <- kendalls_w(mrt ~ interaction(condition, congruency) | pno, data = dat)
+    W <- kendalls_w(mrt ~ interaction(condition, congruency) | pno, data = dat, verbose = FALSE)
     expect_equal(W[[1]], 0.4011, tolerance = 0.01)