fixes, docs

Author: strengejacke

R/collapse_by_group.R

@@ -34,7 +34,7 @@ collapse_by_group <- function(grid, model, collapse_by = NULL, residuals = FALSE
 
   model_data <- insight::get_data(model, source = "frame", verbose = FALSE)
 
-  if (is.null(collapse_by) || isTRUE(collapse_by)) {
+  if (is.null(collapse_by) || isTRUE(residuals)) {
     collapse_by <- insight::find_random(model, flatten = TRUE)
   }
 
@@ -47,15 +47,15 @@ collapse_by_group <- function(grid, model, collapse_by = NULL, residuals = FALSE
   }
 
   if (!collapse_by %in% colnames(model_data)) {
-    insight::format_error("Could not find `", collapse_by, "` column.")
+    insight::format_error(paste0("Could not find `", collapse_by, "` column."))
   }
 
   if (residuals) {
     rawdata <- residualize_over_grid(grid, model)
     y_name <- "Mean"
   } else {
     rawdata <- insight::get_data(model, source = "environment", verbose = FALSE)
-    y_name <- attributes(grid)$response
+    y_name <- insight::find_response(model)
 
     # we need this column for labelling data points, but not for collapsing
     rawdata$rowname <- NULL
@@ -83,6 +83,10 @@ collapse_by_group <- function(grid, model, collapse_by = NULL, residuals = FALSE
     FUN = mean
   )
 
+  if (residuals) {
+    y_name <- insight::find_response(model)
+  }
+
   colnames(agg_data)[ncol(agg_data)] <- y_name
   colnames(agg_data)[colnames(agg_data) == "group"] <- "group_col"
 

R/visualisation_recipe.R

@@ -23,8 +23,10 @@
 #'
 #' @param x A modelbased object.
 #' @param show_data Logical, if `TRUE`, display the "raw" data as a background
-#' to the model-based estimation. This argument will be ignored for plotting
-#' objects returned by `estimate_slopes()` or `estimate_grouplevel()`.
+#' to the model-based estimation. For mixed models, you can additional use the
+#' `collapse_group` argument to "collapse" data points by random effects
+#' grouping factors. Argument `show_data` will be ignored for plotting objects
+#' returned by `estimate_slopes()` or `estimate_grouplevel()`.
 #' @param join_dots Logical, if `TRUE` (default) and for categorical focal terms
 #' in `by`, dots (estimates) are connected by lines, i.e. plots will be a
 #' combination of dots with error bars and connecting lines. If `FALSE`, only
@@ -39,14 +41,17 @@
 #' predictor. Use `FALSE` to always use continuous color scales for numeric
 #' predictors. It is possible to set a global default value using `options()`,
 #' e.g. `options(modelbased_numeric_as_discrete = 10)`.
-#' @param show_residuals Logical, if `TRUE`, display residuals of the model
-#' as a background to the model-based estimation. Residuals will be computed
-#' for the predictors in the data grid, using [`residualize_over_grid()`].
-#' @param collapse_group For mixed effects models, name of the grouping variable
-#' of random effects. If `collapse_group = TRUE`, data points "collapsed" by the
-#' first random effect groups are added to the plot. Else, if `collapse_group`
-#' is a name of a group factor, data is collapsed by that specific random
-#' effect. See [`collapse_by_group()`] for further details.
+#' @param show_residuals Logical, if `TRUE`, display residuals of the model as a
+#' background to the model-based estimation. Residuals will be computed for the
+#' predictors in the data grid, using [`residualize_over_grid()`]. For mixed
+#' models, you can additional use the `collapse_group` argument to "collapse"
+#' data points from residuals by random effects grouping factors.
+#' @param collapse_group This argument only takes effect when either `show_data`
+#' or `show_residuals` is `TRUE`. For mixed effects models, name of the grouping
+#' variable of random effects. If `collapse_group = TRUE`, data points
+#' "collapsed" by the first random effect groups are added to the plot. Else, if
+#' `collapse_group` is a name of a group factor, data is collapsed by that
+#' specific random effect. See [`collapse_by_group()`] for further details.
 #' @param point,line,pointrange,ribbon,facet,grid Additional
 #' aesthetics and parameters for the geoms (see customization example).
 #' @param ... Arguments passed from `plot()` to `visualisation_recipe()`, or

R/visualisation_recipe_internal.R

@@ -322,7 +322,7 @@
   }
 
   # add raw data as first layer ----------------------------------
-  if (show_data) {
+  if (show_data && is.null(collapse_by)) {
     layers[[paste0("l", l)]] <- .visualization_recipe_rawdata(x, aes, numeric_as_discrete)
     # Update with additional args
     if (!is.null(point)) {
@@ -332,7 +332,7 @@
   }
 
   # add residual data as next lowest layer
-  if (show_residuals) {
+  if (show_residuals && is.null(collapse_by)) {
     layers[[paste0("l", l)]] <- .visualization_recipe_residuals(
       x,
       aes,
@@ -346,12 +346,13 @@
   }
 
   # next possible data layer is collapsing by group ---------------
-  if (!is.null(collapse_by)) {
+  if (!is.null(collapse_by) && (show_data || show_residuals)) {
     layers[[paste0("l", l)]] <- .visualization_recipe_collapsed_data(
       x,
       aes,
       collapse_by,
-      numeric_as_discrete
+      numeric_as_discrete,
+      show_residuals
     )
     # Update with additional args
     if (!is.null(point)) {
@@ -597,17 +598,16 @@
   x,
   aes,
   collapse_by,
-  numeric_as_discrete = 8
+  numeric_as_discrete = 8,
+  show_residuals = FALSE
 ) {
   model <- attributes(x)$model
-  grid <- attributes(x)$datagrid
-
-  rawdata <- collapse_by_group(grid, model, collapse_by)
+  rawdata <- collapse_by_group(x, model, collapse_by, residuals = show_residuals)
 
   # Add response to data if not there
-  y <- insight::find_response(attributes(x)$model)
+  y <- insight::find_response(model)
   if (!y %in% names(rawdata)) {
-    rawdata[y] <- insight::get_response(attributes(x)$model, verbose = FALSE)
+    rawdata[y] <- insight::get_response(model, verbose = FALSE)
   }
 
   # if we have less than 8 values for the legend, a continuous color scale
@@ -626,13 +626,20 @@
     )
   }
 
-  .data_point_geom(model = model, aes = aes, data = rawdata, y = y)
+  .data_point_geom(
+    model = model,
+    aes = aes,
+    data = rawdata,
+    y = y,
+    size = 2.5,
+    jitter = 0.1
+  )
 }
 
 
 # helpers -----------------------------------------------------------------
 
-.data_point_geom <- function(model, aes, data, y) {
+.data_point_geom <- function(model, aes, data, y, ...) {
   if (aes$type == "pointrange" && !is.numeric(data[[aes$x]])) {
     geom <- "jitter"
   } else {
@@ -656,6 +663,15 @@
     stroke = stroke
   )
 
+  dots <- list(...)
+  if (!is.null(dots$size)) {
+    out$size <- dots$size
+  }
+  if (!is.null(dots$jitter)) {
+    out$position <- "jitter"
+    out$width <- dots$jitter
+  }
+
   # check if we have matching columns in the raw data - some functions,
   # likes slopes, have mapped these aes to other columns that are not part
   # of the raw data - we set them to NULL