update effectsize, remove chen2010 test, update snapshots (#473)

* update effectsize, remove chen2010 test, update snapshots

* update word list, devel versions for see, performance, bayestestR

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: report
 Type: Package
 Title: Automated Reporting of Results and Statistical Models
-Version: 0.5.9.1
+Version: 0.5.9.2
 Authors@R:
     c(person(given = "Dominique",
              family = "Makowski",
@@ -154,4 +154,4 @@ Collate:
     'utils_misspelled_variables.R'
     'zzz.R'
 Roxygen: list(markdown = TRUE)
-Remotes: easystats/insight
+Remotes: easystats/insight, easystats/effectsize, easystats/see, easystats/performance, easystats/bayestestR

NEWS.md

@@ -1,3 +1,8 @@
+# report 0.6.0
+
+Minor changes
+* `report_htest_chi2` stops supporting rule "chen2010" (following change in `effectsize`).
+
 # report 0.5.9
 
 Breaking
@@ -9,7 +14,7 @@ Breaking
   * `report_participants()`
   * `report_sample()`
 
-Minor changes 
+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`.

R/report_htest_chi2.R

@@ -98,11 +98,16 @@ chi2_type <- function(x) {
 # report_parameters ----------------------------
 
 .report_parameters_chi2 <- function(table, stats, effsize, ...) {
+  if (is.null(attributes(effsize)$interpretation)) {
+    and <- ""
+  } else {
+    and <- paste0(", and ", attributes(effsize)$interpretation)
+  }
+
   text_full <- paste0(
     "statistically ",
     effectsize::interpret_p(table$p, rules = "default"),
-    ", and ",
-    attributes(effsize)$interpretation,
+    and,
     " (",
     stats,
     ")"
@@ -111,8 +116,7 @@ chi2_type <- function(x) {
   text_short <- paste0(
     "statistically ",
     effectsize::interpret_p(table$p, rules = "default"),
-    ", and ",
-    attributes(effsize)$interpretation,
+    and,
     " (",
     summary(stats),
     ")"

inst/WORDLIST

@@ -4,7 +4,6 @@ BayesFactor
 BibLaTeX
 CMD
 CSL
-Dom
 ELPD
 ENP
 ESS
@@ -25,26 +24,21 @@ Sivula
 Vehtari
 amongst
 anova
-bmwiernik
 brms
+chen
 eXistence
 easystats
 elpd
 github
 htest
 ivreg
-lifecycle
-mattansb
 pacakges
 participants’
-patilindrajeets
-rempsyc
 repo
 reproducibility
 rstanarm
 sIgnificance
 setosa
-strengejacke
 tidyverse
 unarchive
 versicolor

tests/testthat/_snaps/windows/report.htest-chi2.md

@@ -143,18 +143,6 @@
       statistically significant, and medium (chi2 = 31.57, p < .001; Cohen's h =
       0.74, 95% CI [0.50, 0.99])
 
----
-
-    Code
-      report(x, type = "oddsratio", rules = "chen2010")
-    Output
-      Effect sizes were labelled following Chen's (2010) recommendations.
-      
-      The Pearson's Chi-squared test with Yates' continuity correction of
-      independence between Diagnosis and Group suggests that the effect is
-      statistically significant, and medium (chi2 = 31.57, p < .001; Odds ratio =
-      4.73, 95% CI [2.74, 8.17])
-
 ---
 
     Code
@@ -164,8 +152,8 @@
       
       The Pearson's Chi-squared test with Yates' continuity correction of
       independence between Diagnosis and Group suggests that the effect is
-      statistically significant, and (chi2 = 31.57, p < .001; Risk_ratio = 2.54, 95%
-      CI [1.80, 3.60])
+      statistically significant (chi2 = 31.57, p < .001; Risk_ratio = 2.54, 95% CI
+      [1.80, 3.60])
 
 ---
 

tests/testthat/test-report.htest-chi2.R

@@ -81,11 +81,6 @@ test_that("report.htest-chi2 report", {
     report(x, type = "cohens_h", rules = "sawilowsky2009")
   )
 
-  expect_snapshot(
-    variant = "windows",
-    report(x, type = "oddsratio", rules = "chen2010")
-  )
-
   expect_snapshot(
     variant = "windows",
     report(x, type = "riskratio")