0.2.4

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: manymome
 Title: Mediation, Moderation and Moderated-Mediation After Model Fitting
-Version: 0.2.3.14
+Version: 0.2.4
 Authors@R:
     c(person(given = "Shu Fai",
            family = "Cheung",

NEWS.md

@@ -1,4 +1,4 @@
-# manymome 0.2.3.14
+# manymome 0.2.4
 
 ## New Features
 
@@ -54,9 +54,19 @@
 
 - Added `wrap_computation` to the argument
   of `print.indirect()` to wrap the printout
-  of long computational formulas.
+  of long computational formulas. Default
+  set to `TRUE`.
   (0.2.3.14)
 
+- Updated `plot_effect_vs_w()` to check
+  whether a moderator is numeric.
+  (0.2.3.15)
+
+- Updated `print.cond_indirect_effects()`.
+  Disabled the printing of the (only)
+  component path when a path has no
+  mediator. (0.2.3.16)
+
 ## Documentation and Notes
 
 - Corrected `README` to remark that

R/plot_effect_vs_w.R

@@ -314,7 +314,7 @@ plot_effect_vs_w <- function(object,
     if (ncol(wlevels) > 1) {
         stop("Only an effect with one moderator is supported.")
       }
-    if (!is.numeric(wlevels[, 1])) {
+    if (!((isTRUE(w_type == "numeric")) || is.null(w_type))) {
         stop("Only numeric moderators are supported.")
       }
     if (is.null(w)) {

R/print_cond_indirect_effect.R

@@ -390,9 +390,13 @@ print.cond_indirect_effects <- function(x, digits = 3,
       i <- which(names(out) == "ind")
       j <- length(out)
       out <- c(out[1:i], out_original, out[(i + 1):j])
-      # Drop the component column
+    }
+  # Drop the component column if the model has no mediator
+  if (!has_m) {
       i <- which(names(out) %in% names(x_i$components))
-      out <- out[-i]
+      if (length(i) > 0) {
+          out <- out[-i]
+        }
     }
   out1 <- data.frame(out, check.names = FALSE)
   if (has_wlevels) {
@@ -557,7 +561,7 @@ print.cond_indirect_effects <- function(x, digits = 3,
                 ifelse(has_wlevels,
                        "the moderator(s).",
                        "the group(s)."))
-      if (!print_original_se) {
+      if (has_m) {
           cat(strwrap(paste("\n -", paste(sQuote(mcond), collapse = ","),
               "is/are the path coefficient(s) along the path",
               "conditional on",

README.md

@@ -9,7 +9,7 @@
 [![R-CMD-check](https://github.com/sfcheung/manymome/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/sfcheung/manymome/actions/workflows/R-CMD-check.yaml)
 <!-- badges: end -->
 
-(Version 0.2.3.14, updated on 2024-10-02, [release history](https://sfcheung.github.io/manymome/news/index.html))
+(Version 0.2.4, updated on 2024-10-04, [release history](https://sfcheung.github.io/manymome/news/index.html))
 
 # manymome  <img src="man/figures/logo.png" align="right" height="150" />
 

tests/testthat/test_mod_only_plot_vs_w.R

@@ -69,8 +69,9 @@ sos_2 <- cond_indirect_effects(wlevels = "w1",
 sos_3 <- cond_indirect_effects(wlevels = c("w1", "w2"),
                                x = "m", y = "y",
                                fit = fit3)
-
-plot_effect_vs_w(cos_1a, level = .99) + ggplot2::ylim(0, 1)
-plot_effect_vs_w(cos_1a) + ggplot2::ylim(0, 1)
-plot_effect_vs_w(cos_1a, level = .60) + ggplot2::ylim(0, 1)
-plot_effect_vs_w(cos_1a, level = .50) + ggplot2::ylim(0, 1)
+test_that("mod only", {
+expect_no_error(plot_effect_vs_w(cos_1a, level = .99) + ggplot2::ylim(0, 1))
+expect_no_error(plot_effect_vs_w(cos_1a) + ggplot2::ylim(0, 1))
+expect_no_error(plot_effect_vs_w(cos_1a, level = .60) + ggplot2::ylim(0, 1))
+expect_no_error(plot_effect_vs_w(cos_1a, level = .50) + ggplot2::ylim(0, 1))
+})

tests/testthat/test_plot_indirect_vs_w.R

@@ -240,6 +240,15 @@ test_that("Expand wlevels", {
   p1
 })
 
+test_that("Detect categorical moderators", {
+dat <- data_med_mod_serial_cat
+lm_m <- lm(m1 ~ x*w2, dat)
+lm_y <- lm(y ~ m1 + x, dat)
+lm_all <- lm2list(lm_m, lm_y)
+cond_out <- cond_indirect_effects(x = "x", y = "y", m = "m1", wlevels = "w2", fit = lm_all)
+expect_error(plot_effect_vs_w(cond_out), "numeric moderators")
+})
+
 # Multigroup
 
 # On hold. cond_indirect_effects() does not yet support

vignettes/articles/med_mg.Rmd

@@ -1,7 +1,7 @@
 ---
 title: "Multigroup Models With Mediation Effects"
 author: "Shu Fai Cheung & Sing-Hang Cheung"
-date: "2024-09-25"
+date: "2024-10-04"
 output:
   html_document:
     fig.align: "center"
@@ -85,7 +85,7 @@ These are the estimates of the paths:
 ``` r
 summary(fit,
         estimates = TRUE)
-#> lavaan 0.6-18 ended normally after 1 iteration
+#> lavaan 0.6-19 ended normally after 1 iteration
 #> 
 #>   Estimator                                         ML
 #>   Optimization method                           NLMINB
@@ -204,7 +204,7 @@ fit_boot_out <- do_boot(fit = fit,
                         seed = 53253,
                         ncores = 8)
 #> 8 processes started to run bootstrapping.
-#> The expected CPU time is about 22.5 second(s).
+#> The expected CPU time is about 15 second(s).
 ```
 
 # Estimate Indirect Effects
@@ -253,9 +253,11 @@ ind_gpA
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.87989)*(0.46481)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 5000 bootstrap samples.
 #> 
@@ -304,9 +306,11 @@ ind_gpB
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.59716)*(1.11040)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 5000 bootstrap samples.
 #> 
@@ -512,9 +516,11 @@ std_gpA
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)*sd_x/sd_y
+#> 
 #> Computation:
 #>   (0.87989)*(0.46481)*(1.10557)/(2.21581)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 5000 bootstrap samples.
 #> 
@@ -553,9 +559,11 @@ std_gpB
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)*sd_x/sd_y
+#> 
 #> Computation:
 #>   (0.59716)*(1.11040)*(1.00943)/(2.58386)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 5000 bootstrap samples.
 #> 
@@ -825,15 +833,14 @@ path1
 #>  Path: m -> y
 #>  Conditional on group(s): Group A[1], Group B[2]
 #> 
-#>     Group Group_ID   ind CI.lo CI.hi Sig   y~m
-#> 1 Group A        1 0.465 0.110 0.819 Sig 0.465
-#> 2 Group B        2 1.110 0.765 1.475 Sig 1.110
+#>     Group Group_ID   ind CI.lo CI.hi Sig
+#> 1 Group A        1 0.465 0.110 0.819 Sig
+#> 2 Group B        2 1.110 0.765 1.475 Sig
 #> 
 #>  - [CI.lo to CI.hi] are 95.0% percentile confidence intervals by
 #>    nonparametric bootstrapping with 5000 samples.
 #>  - The 'ind' column shows the direct effects.
-#>  - 'y~m' is/are the path coefficient(s) along the path conditional on
-#>    the group(s).
+#> 
 ```
 
 The difference between the two paths can be tested
@@ -852,9 +859,9 @@ path1_diff
 #>  Path: m -> y
 #>  Conditional on group(s): Group B[2], Group A[1]
 #> 
-#>     Group Group_ID   ind CI.lo CI.hi Sig   y~m
-#> 1 Group B        2 1.110 0.765 1.475 Sig 1.110
-#> 2 Group A        1 0.465 0.110 0.819 Sig 0.465
+#>     Group Group_ID   ind CI.lo CI.hi Sig
+#> 1 Group B        2 1.110 0.765 1.475 Sig
+#> 2 Group A        1 0.465 0.110 0.819 Sig
 #> 
 #> == Difference in Conditional Indirect Effect ==
 #> 

vignettes/do_boot.Rmd

@@ -1,7 +1,7 @@
 ---
 title: "Generate Bootstrap Estimates"
 author: "Shu Fai Cheung & Sing-Hang Cheung"
-date: "2024-09-25"
+date: "2024-10-04"
 output:
   rmarkdown::html_vignette:
     number_sections: true
@@ -86,7 +86,7 @@ fit_lavaan <- sem(model = mod, data = dat,
            fixed.x = FALSE,
            estimator = "MLR")
 summary(fit_lavaan)
-#> lavaan 0.6-18 ended normally after 1 iteration
+#> lavaan 0.6-19 ended normally after 1 iteration
 #> 
 #>   Estimator                                         ML
 #>   Optimization method                           NLMINB
@@ -244,9 +244,11 @@ out_lavaan
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.93469)*(0.78469)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 500 bootstrap samples.
 #> 
@@ -344,7 +346,7 @@ regression outputs to one object using
 fit_lm <- lm2list(lm_m, lm_y)
 fit_lm
 #> 
-#> The models:
+#> The model(s):
 #> m ~ x + c1 + c2
 #> y ~ m + x + c1 + c2
 ```
@@ -402,9 +404,11 @@ out_lm
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.93469)*(0.78469)
 #> 
+#> 
 #> Percentile confidence interval formed by nonparametric bootstrapping
 #> with 500 bootstrap samples.
 #> 

vignettes/do_mc.Rmd

@@ -1,7 +1,7 @@
 ---
 title: "Generate Monte Carlo Estimates"
 author: "Shu Fai Cheung & Sing-Hang Cheung"
-date: "2024-09-25"
+date: "2024-10-04"
 output:
   rmarkdown::html_vignette:
     number_sections: true
@@ -106,7 +106,7 @@ fit_lavaan <- sem(model = mod, data = dat,
                   fixed.x = FALSE,
                   estimator = "MLR")
 summary(fit_lavaan)
-#> lavaan 0.6-18 ended normally after 1 iteration
+#> lavaan 0.6-19 ended normally after 1 iteration
 #> 
 #>   Estimator                                         ML
 #>   Optimization method                           NLMINB
@@ -225,9 +225,11 @@ out_lavaan
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.93469)*(0.78469)
 #> 
+#> 
 #> Monte Carlo confidence interval with 10000 replications.
 #> 
 #> Coefficients of Component Paths:

vignettes/do_mc_lavaan_mi.Rmd

@@ -1,7 +1,7 @@
 ---
 title: "Monte Carlo Confidence Intervals with Multiple Imputation"
 author: "Shu Fai Cheung & Sing-Hang Cheung"
-date: "2024-09-25"
+date: "2024-10-04"
 output:
   rmarkdown::html_vignette:
     number_sections: true
@@ -246,9 +246,11 @@ out_lavaan
 #> 
 #> Computation Formula:
 #>   (b.m~x)*(b.y~m)
+#> 
 #> Computation:
 #>   (0.89141)*(0.73569)
 #> 
+#> 
 #> Monte Carlo confidence interval with 10000 replications.
 #> 
 #> Coefficients of Component Paths: