Skip to contents

Visualise Health Economic Model Datasets

Source: vignettes/V_04.Rmd
V_04.Rmd

Note: This vignette is illustrated with fake data. The dataset explored in this example should not be used to inform decision-making.

ready4use includes a number of tools for visualising health economic model data and forms part of the ready4 framework. Details of how to find compatible datasets are provided in another article. The ready4use visualisation tools illustrated in this vignette provide an interface to functions from the ggpubr library. To work as intended, some (but not most) of these tools also require the ggpubr library to be loaded.

## Loading required package: ggplot2

Ingest and pre-process data

We begin by ingesting the data from online repository we require into a Ready4useDyad, as illustrated in another vignette.

objects_ls <- Ready4useRepos(dv_nm_1L_chr = "fakes",
                             dv_ds_nm_1L_chr = "https://doi.org/10.7910/DVN/HJXYKQ",
                             dv_server_1L_chr = "dataverse.harvard.edu") %>%
  ingest(fls_to_ingest_chr = c("ymh_clinical_tb","ymh_clinical_dict_r3"),
         metadata_1L_lgl = F)
X <- Ready4useDyad(ds_tb = objects_ls$ymh_clinical_tb,
                   dictionary_r3 = objects_ls$ymh_clinical_dict_r3) %>% 
  renew(type_1L_chr = "case")

This dataset is a synthetic (“fake”) microdata representation of clinic patients with data for each patient reported for up to two data collection rounds (baseline and follow-up). We also need to transform this dataset into other formats to use with some of the plot types we illustrate in this vignette. We first make a dataset with baseline values only.

X1 <- renewSlot(X, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::filter(round=="Baseline"))

Next, we create summary datasets with mean variable values for the entire sample at baseline.

X2 <- renewSlot(X1, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::group_by(d_studying_working) %>%
                  dplyr::summarise(dplyr::across(dplyr::where(is.numeric), 
                                                 function(x){ mean(x, na.rm = TRUE)})))
X3 <- renewSlot(X1, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::group_by(d_studying_working, d_sex_birth_s) %>%
                  dplyr::summarise(dplyr::across(dplyr::where(is.numeric), 
                                                 function(x){ mean(x, na.rm = TRUE)}), .groups = 'drop'))
X4 <- renewSlot(X1, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::group_by(d_sex_birth_s, d_country_bir_s, 
                                                            d_studying_working, c_p_diag_s,c_clinical_staging_s) %>%
                  dplyr::summarise(dplyr::across(dplyr::where(is.numeric), 
                                                 function(x){ mean(x, na.rm = TRUE)}), .groups = 'drop'))

We also create datasets with summaries of mean variable values at both timepoints.

X5 <- renewSlot(X, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::group_by(round) %>%
                  dplyr::summarise(dplyr::across(dplyr::where(is.numeric), function(x){ mean(x, na.rm = TRUE)})))
X6 <- renewSlot(X, "ds_tb",
                procureSlot(X, "ds_tb") %>% dplyr::group_by(round, d_sex_birth_s) %>%
                  dplyr::summarise(dplyr::across(dplyr::where(is.numeric), function(x){ mean(x, na.rm = TRUE)})))
## `summarise()` has grouped output by 'round'. You can override using the
## `.groups` argument.

We make a dataset that is restricted to 50 randomly selected cases for which data is available at two timepoints.

X7 <- renewSlot(X, "ds_tb", 
                procureSlot(X, "ds_tb") %>% 
                  dplyr::filter(fkClientID %in% (intersect(procureSlot(X, "ds_tb") %>% dplyr::filter(round=="Follow-up") %>%
                                                             dplyr::pull(fkClientID), X1@ds_tb$fkClientID) %>% sample(50))))

We next make dataset summaries of the counts at baseline of sample sub-groups.

X8 <- renewSlot(X1, "ds_tb", 
                procureSlot(X1, "ds_tb") %>% dplyr::filter(!is.na(d_studying_working)) %>% 
                  dplyr::group_by(d_studying_working) %>% dplyr::summarise(Count = dplyr::n()))
X9 <- renewSlot(X1, "ds_tb", 
                table(procureSlot(X1, "ds_tb") %>% dplyr::select(d_studying_working, d_sex_birth_s)) %>% tibble::as_tibble())
X10 <- renewSlot(X1, "ds_tb", 
                 table(procureSlot(X1, "ds_tb") %>%
                         dplyr::select(d_studying_working, d_sex_birth_s, c_p_diag_s, c_clinical_staging_s)) %>% tibble::as_tibble())

Some of the visualisations that we will be generating in this vignette will require labels that are derived from variable values to be modified (e.g. for brevity). We therefore next make look-up tables for recoding these values.

x <- renew(ready4show::ready4show_correspondences(),
      old_nms_chr = c("Not studying or working", "Studying only", "Studying and working", "Working only", "Female", "Male"),
      new_nms_chr = c("NEITHER", "EDUCATION", "BOTH", "EMPLOYMENT", "FEMALE", "MALE"))
y <- renew(ready4show::ready4show_correspondences(),
           old_nms_chr = c("Not studying or working", "Studying only", "Studying and working", "Working only", 
                           "Depression and Anxiety"),
           new_nms_chr = c("Neither", "Education", "Both", "Employment", "Both"))

Options for using the Depict method illustrated with the default plot type (barplot)

Data contained in a Ready4useDyad can be visualised using the Depict method.

Single plots

Plot results for one variable

Basic plot 
depict(X1, x_vars_chr = "d_studying_working")

Drop missing values 
depict(X1, x_vars_chr = "d_studying_working", drop_missing_1L_lgl = T)

Drop tick marks and tick lables 
depict(X1, x_vars_chr = "d_studying_working", drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T)

Use variable description instead of variable name (via lookup to data dictionary) for X axis and legend labels and remove label from Y axis 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, y_labels_chr = "", 
                     z_labels_chr = NA_character_, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T)
## Ignoring unknown labels:
## • shape : "Education and employment status"
## • linetype : "Education and employment status"

Specify custom X axis and legend labels 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
                     z_labels_chr = "", drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Change Y axis from counts to percentage 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
                     z_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Recode the legend value labels 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
       z_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T,
       recode_lup_r3 = x)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Stratify results by a second (categorical) variable - in this case, sex 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
       z_vars_chr = "d_sex_birth_s", z_labels_chr = "SEX", as_percent_1L_lgl = T, drop_missing_1L_lgl = T,
       recode_lup_r3 = x)
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"

Stack bars 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
       z_vars_chr = "d_sex_birth_s", z_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T,
       position_xx = ggplot2::position_stack(), recode_lup_r3 = y)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Plot results for two variables - one categorical (x axis) and the other numeric (y axis)

Generate title using variable labels 
depict(X2, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, y_vars_chr = "k6_total",
       y_labels_chr = NA_character_, z_labels_chr = "", drop_missing_1L_lgl = T, 
       drop_ticks_1L_lgl = T, recode_lup_r3 = x, titles_chr = NA_character_)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Stratify results by a third (categorical) variable - in this case, sex 
depict(X3, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_vars_chr = "k6_total",
                     y_labels_chr = "K6", z_vars_chr = "d_sex_birth_s", z_labels_chr = "SEX", drop_missing_1L_lgl = T,
                     titles_chr = "PSYCHOLOGICAL DISTRESS BY EDUCATION, EMPLOYMENT AND SEX")
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"

Modify palettes

The palette options from the “ggsci” package that are compatible with the ready4use Depict method can be retrieved using the following function call.

get_styles("ggsci")
##  [1] "npg"          "aaas"         "lancet"       "jco"          "nejm"        
##  [6] "ucscgb"       "uchicago"     "d3"           "futurama"     "igv"         
## [11] "locuszoom"    "rickandmorty" "startrek"     "simpsons"     "tron"
Use Lancet journal palette 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
                     z_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T, style_1L_chr = "lancet")
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

The palette options from “the viridis” package that are compatible with the ready4use Depict method can be retrieved using the following function call.

get_styles("viridis")
##  [1] "magma"   "A"       "inferno" "B"       "plasma"  "C"       "viridis"
##  [8] "D"       "cividis" "E"       "rocket"  "F"       "mako"    "G"      
## [15] "turbo"   "H"
Use the Viridis Rocket palette 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
       z_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T, 
       style_1L_chr = "rocket", type_1L_chr = "viridis")
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Use a custom palette 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", y_labels_chr = "",
       z_labels_chr = "", as_percent_1L_lgl = T, colours_chr = c("#130d9c","#9c0d39"), drop_missing_1L_lgl = T,
       drop_ticks_1L_lgl = T, type_1L_chr = "manual")
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Single colour plots

Use just the first colour from the selected (in this case, default) palette 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x, 
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS")

Single colour plot using AAAS journal palette 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x, style_1L_chr = "aaas", 
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS")

Single colour plot using custom colour (black) 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       colours_chr = "black",  drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x,
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS", type_1L_chr = "manual")

Supply additional ggpubr arguments

Add label to bars 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x, 
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS", label = T, lab.pos = "in", lab.col = "white")

Change orientation of bars 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x, 
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS", orientation = "horiz")

Change order of bars 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", y_labels_chr = "", as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, fill_single_1L_lgl = T, recode_lup_r3 = x, 
       titles_chr = "EDUCATION AND EMPLOYMENT STATUS", order = c("EDUCATION", "EMPLOYMENT","BOTH", "NEITHER"))

Create multiple plots

Create a list of plots 

plot_ls <- depict(X1, x_vars_chr = c("d_sex_birth_s","d_studying_working", "d_country_bir_s"), 
                  x_labels_chr = NA_character_, y_labels_chr = "" , z_labels_chr = "", 
                  as_percent_1L_lgl = T, drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T)
plot_ls$d_sex_birth_s
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

plot_ls$d_studying_working
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

plot_ls$d_country_bir_s
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Create a composite plot, with a common legend 

depict(X4, x_vars_chr = c("d_country_bir_s", "d_studying_working", "c_p_diag_s", "c_clinical_staging_s"),
       x_labels_chr = NA_character_, y_vars_chr = "k6_total", y_labels_chr = "K6", 
       z_vars_chr = "d_sex_birth_s", z_labels_chr = "SEX", arrange_1L_lgl = T,  
       arrange_args_ls = list(ncol = 2, nrow = 2, common.legend = T), 
       drop_missing_1L_lgl = T, recode_lup_r3 = y)
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"
## Ignoring unknown labels:
## • shape : "SEX"
## • linetype : "SEX"

Other plot types

Plot one continuous variable

Density plot

Basic density plot 
depict(X1, x_vars_chr = "k6_total", what_1L_chr = "density")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_density()`).

Stratify results by a second (categorical) variable, report percentages rather than probability and look-up labels from dictionary 
depict(X1, x_vars_chr = "k6_total", x_labels_chr = NA_character_, y_labels_chr = "Percentatge", 
       z_vars_chr = "d_studying_working", z_labels_chr = "",  as_percent_1L_lgl = T,
       drop_missing_1L_lgl = T, what_1L_chr = "density") 
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Empirical cumulative density function (ECDF)

Basic ECDF plot 
depict(X1, x_vars_chr = "k6_total", what_1L_chr = "ecdf")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_ecdf()`).

Stratify results by a second (categorical) variable and customise labels 
depict(X1,x_vars_chr = "k6_total", x_labels_chr = NA_character_, y_labels_chr = "", 
       z_vars_chr = "d_studying_working", z_labels_chr = "", drop_missing_1L_lgl = T,  
       what_1L_chr = "ecdf")
## Ignoring unknown labels:
## • fill : ""
## • shape : ""

Histogram

Basic histogram 
depict(X1, x_vars_chr = "k6_total", what_1L_chr = "histogram")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_bin()`).

Specify the number of bins, report percentages rather than counts and customise labels 
depict(X1, x_vars_chr = "k6_total", x_labels_chr = "K6", y_labels_chr = "", as_percent_1L_lgl = T,
                     what_1L_chr = "histogram", bins = 10)
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_bin()`).

Stratify results by a second (categorical) variable and look-up labels from dictionary 
depict(X1, x_vars_chr = "k6_total", x_labels_chr = NA_character_, z_vars_chr = "d_sex_birth_s",
       z_labels_chr = NA_character_, as_percent_1L_lgl = T, drop_missing_1L_lgl = T, position_xx = "dodge",
       what_1L_chr = "histogram", bins=10)
## Ignoring unknown labels:
## • shape : "Sex at birth"
## • linetype : "Sex at birth"

Add custom ggpubr arguments to add density plot and mean lines 
depict(X1, x_vars_chr = "k6_total", x_labels_chr = "K6", y_labels_chr = "", 
       z_vars_chr = "d_sex_birth_s", z_labels_chr = "Sex", as_percent_1L_lgl = F, 
       drop_missing_1L_lgl = T, what_1L_chr = "histogram", add = "mean", add_density = TRUE)
## Ignoring unknown labels:
## • shape : "Sex"
## • linetype : "Sex"

QQ Plot

Basic QQ plot 
depict(X1, x_vars_chr = "k6_total", what_1L_chr = "qqplot")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_qq()`).
## Warning: Removed 1 row containing non-finite outside the scale range (`stat_qq_line()`).
## Removed 1 row containing non-finite outside the scale range (`stat_qq_line()`).

Stratify results by a second (categorical) variable and customise labels 
depict(X1, x_vars_chr = "k6_total",  y_labels_chr = "K6", drop_missing_1L_lgl = T,
       z_vars_chr = "d_studying_working", z_labels_chr = "", what_1L_chr = "qqplot")
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Plot two variables (one continuous, one discrete)

Box plot

Basic box plot 
depict(X, x_vars_chr = "round", y_vars_chr = "k6_total", what_1L_chr = "boxplot")
## Warning: Removed 3 rows containing non-finite outside the scale range
## (`stat_boxplot()`).

Customise labels and remove x-axis tick-marks 
depict(X, x_vars_chr = "round", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = "K6", z_labels_chr = "", drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T, 
       what_1L_chr = "boxplot")
## Ignoring unknown labels:
## • fill : ""
## • shape : ""
## • linetype : ""

Fill boxes 
depict(X, x_vars_chr = "round", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = NA_character_, z_labels_chr = "", drop_missing_1L_lgl = T, drop_ticks_1L_lgl = T, 
       what_1L_chr = "boxplot", fill = "round", fill_single_1L_lgl = F)
## Ignoring unknown labels:
## • shape : ""
## • linetype : ""

Stratify results by a third (categorical) variable 
depict(X, x_vars_chr = "round", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = NA_character_, drop_missing_1L_lgl = T, z_vars_chr = "d_sex_birth_s", 
       z_labels_chr = NA_character_, what_1L_chr = "boxplot", fill = "d_sex_birth_s", fill_single_1L_lgl = F)
## Ignoring unknown labels:
## • shape : "Sex at birth"
## • linetype : "Sex at birth"

Add jitter points 
depict(X, x_vars_chr = "round", x_labels_chr = "", y_vars_chr = "k6_total", y_labels_chr = "K6", 
       drop_missing_1L_lgl = T, z_vars_chr = "d_sex_birth_s", z_labels_chr = NA_character_, 
       what_1L_chr = "boxplot", add = "jitter")
## Ignoring unknown labels:
## • fill : "Sex at birth"
## • shape : "Sex at birth"
## • linetype : "Sex at birth"

Donut chart

Basic donut chart 
depict(X1, x_vars_chr = "d_studying_working", what_1L_chr = "donutchart")
## Ignoring unknown labels:
## • colour : "d_studying_working"
## • shape : "d_studying_working"
## • linetype : "d_studying_working"

Change colour of separating line, drop missing values and remove label 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", drop_missing_1L_lgl = T,
       line_1L_chr = "white", what_1L_chr = "donutchart")
## Ignoring unknown labels:
## • colour : ""
## • shape : ""
## • linetype : ""

Report percentage rather than counts and change location and colouring of segment labels 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, 
                     line_1L_chr = "white", what_1L_chr = "donutchart", lab.pos = "in", lab.font = "white")
## Ignoring unknown labels:
## • colour : ""
## • shape : ""
## • linetype : ""

The same result can be achieved by supplying a dataset with frequency counts. Note, drop_missing_1L_lgl is ignored if using this method. 

depict(X8, x_vars_chr = "Count", z_vars_chr = "d_studying_working", z_labels_chr = "", as_percent_1L_lgl = T, 
       line_1L_chr = "white", what_1L_chr = "donutchart", lab.pos = "in", lab.font = "white")
## Ignoring unknown labels:
## • colour : ""
## • shape : ""
## • linetype : ""

Remove legend and change segment labels to names rather than values. 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, drop_legend_1L_lgl = T, drop_missing_1L_lgl = T,
                     line_1L_chr = "white", what_1L_chr = "donutchart", label = "d_studying_working")
## Ignoring unknown labels:
## • colour : "Education and employment status"
## • shape : "Education and employment status"
## • linetype : "Education and employment status"

Dot chart

Basic dot chart 
depict(X2, x_vars_chr = "d_studying_working", y_vars_chr = "k6_total", what_1L_chr = "dotchart")

Use a single colour for dots, rotate chart, increase dot-size, customise label orders, recode label values and add and customise segment lines 
depict(X2, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT", 
       y_vars_chr = "k6_total", y_labels_chr = "K6", drop_missing_1L_lgl = T, fill_single_1L_lgl = T, 
       recode_lup_r3 = y, what_1L_chr = "dotchart", add = "segment", 
       add.params = list(color = "lightgray", size = 1.5),  rotate = T, size = 3)
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## ℹ The deprecated feature was likely used in the ggpubr package.
##   Please report the issue at <https://github.com/kassambara/ggpubr/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Stratify by third (categorical) variable. 
depict(X3, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT",
       y_vars_chr = "k6_total", y_labels_chr = "K6", z_vars_chr = "d_sex_birth_s", z_labels_chr = "SEX",
       drop_missing_1L_lgl = T, position_xx = ggplot2::position_dodge(0.3), recode_lup_r3 = x,
       what_1L_chr = "dotchart", add = "segment", add.params = list(color = "lightgray", size = 1.5),
       rotate = T, size = 3)
## Ignoring unknown labels:
## • fill : "SEX"
## • shape : "SEX"
## • linetype : "SEX"

Dot plot

Basic dot plot 
depict(X1, x_vars_chr = "d_studying_working", y_vars_chr = "k6_total", what_1L_chr = "dotplot")
## Bin width defaults to 1/30 of the range of the data. Pick better value with
## `binwidth`.
## Warning: Removed 33 rows containing missing values or values outside the scale range
## (`stat_bindot()`).

Customise and reorder labels, remove missing values and change dot size 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT" ,
       y_vars_chr = "k6_total",  y_labels_chr = "K6", drop_legend_1L_lgl = T, drop_missing_1L_lgl = T,
       recode_lup_r3 = x, what_1L_chr = "dotplot", 
       order = c("EDUCATION", "EMPLOYMENT","BOTH", "NEITHER"), size = 0.3)
## Bin width defaults to 1/30 of the range of the data. Pick better value with
## `binwidth`.

Stratify by a third (categorical) variable 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "EDUCATION AND EMPLOYMENT" ,y_vars_chr = "k6_total",  y_labels_chr = "K6", 
                     z_vars_chr = "d_sex_birth_s", z_labels_chr = "SEX", drop_missing_1L_lgl = T, what_1L_chr = "dotplot", size = 0.35)
## Ignoring unknown labels:
## • colour : "SEX"
## • shape : "SEX"
## • linetype : "SEX"
## Bin width defaults to 1/30 of the range of the data. Pick better value with
## `binwidth`.

Error plot

Basic error plot 
depict(X1, x_vars_chr = "d_studying_working", y_vars_chr = "k6_total", what_1L_chr = "errorplot")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_summary()`).

Customise labels and the positioning and colouring of error lines 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = "K6", z_vars_chr = "d_sex_birth_s", z_labels_chr = NA_character_,
       drop_missing_1L_lgl = T, what_1L_chr = "errorplot",
       palette = "Paired", error.plot = "pointrange", position = ggplot2::position_dodge(0.5))
## Ignoring unknown labels:
## • fill : "Sex at birth"
## • shape : "Sex at birth"
## • linetype : "Sex at birth"

Line plot

Basic line plot 
depict(X5, x_vars_chr = "round", y_vars_chr = "k6_total", what_1L_chr = "line")

Plot mean values with error bars, stratify results by third (categorical) variable and look-up labels from dictionary 
depict(X, x_vars_chr = "round", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = "K6", z_vars_chr = "d_sex_birth_s", z_labels_chr = NA_character_,
       drop_missing_1L_lgl = T, position_xx = ggplot2::position_dodge(0.5), 
       what_1L_chr = "line", add = "mean_se")
## Ignoring unknown labels:
## • fill : "Sex at birth"

Paired plot

Basic paired plot 
depict(X7, x_vars_chr = "round", y_vars_chr = "k6_total", what_1L_chr = "paired")

Customise labels, remove x-axis ticks and labels and change line colour 
depict(X7, x_vars_chr = "round", x_labels_chr = "", y_vars_chr = "k6_total", 
       y_labels_chr = "K6", z_labels_chr = NA_character_, 
       drop_ticks_1L_lgl = T, line_1L_chr = "lightgray", what_1L_chr = "paired")
## Ignoring unknown labels:
## • fill : "Round of data collection"
## • shape : "Round of data collection"
## • linetype : "Round of data collection"

Pie chart

Basic pie chart 
depict(X1, x_vars_chr = "d_studying_working", what_1L_chr = "pie")
## Ignoring unknown labels:
## • colour : "d_studying_working"
## • shape : "d_studying_working"
## • linetype : "d_studying_working"

Report percentage rather than counts, change location and colouring of segment labels and remove labels 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = "", as_percent_1L_lgl = T, drop_missing_1L_lgl = T, 
       line_1L_chr = "white", what_1L_chr = "pie", lab.pos = "in", lab.font = "white")
## Ignoring unknown labels:
## • colour : ""
## • shape : ""
## • linetype : ""

The same result can be achieved by supplying a dataset with frequency counts. Note, the drop_missing_1L_lgl is ignored if using this method. 

depict(X8, x_vars_chr = "Count", z_vars_chr = "d_studying_working", z_labels_chr = "",
       as_percent_1L_lgl = T, drop_missing_1L_lgl = T, line_1L_chr = "white", what_1L_chr = "pie", 
       lab.pos = "in", lab.font = "white")
## Ignoring drop_missing_1L_lgl argument value - this is only used when not directly supplying a frequency table
## Ignoring unknown labels:
## • colour : ""
## • shape : ""
## • linetype : ""

Strip chart

Basic strip chart 
depict(X1, x_vars_chr = "d_studying_working", y_vars_chr = "k6_total", what_1L_chr = "strip")
## Warning: Removed 1 row containing missing values or values outside the scale range
## (`geom_point()`).

Look-up labels, drop legend and add mean and standard deviation lines 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
                     y_labels_chr = "K6", drop_legend_1L_lgl = T, drop_missing_1L_lgl = T, what_1L_chr = "strip", add = "mean_sd")

Violin plot

Basic violin plot 
depict(X1, x_vars_chr = "d_studying_working", y_vars_chr = "k6_total", what_1L_chr = "violin")
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_ydensity()`).

Customise labels, drop legend and add mean and jitter dots 
depict(X1, x_vars_chr = "d_studying_working", x_labels_chr = NA_character_, y_vars_chr = "k6_total",
       y_labels_chr = "", drop_legend_1L_lgl = T, drop_missing_1L_lgl = T, 
       what_1L_chr = "violin", add = "jitter")

For plotting two variables (both continuous)

Scatter plot

Basic scatter plot 
depict(X1, x_vars_chr = "phq9_total", y_vars_chr = "k6_total", what_1L_chr = "scatter")
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

Add regression line, change dot sizes and and customise labels. 
depict(X1, x_vars_chr = "phq9_total", x_labels_chr = NA_character_, y_vars_chr = "k6_total", 
       y_labels_chr = "K6", drop_missing_1L_lgl = T, what_1L_chr = "scatter", 
       add = "reg.line", size = 0.5, conf.int = TRUE, cor.coef = TRUE, 
       cor.coeff.args = list(method = "pearson", label.x = 3, label.sep = "\n"))

Scatter plot with histograms

Basic scatter plot with histograms 
depict(X1, x_vars_chr = "phq9_total", y_vars_chr = "k6_total", what_1L_chr = "scatterhist")
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning: Removed 4 rows containing non-finite outside the scale range
## (`stat_density()`).
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_density()`).

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning: Removed 4 rows containing non-finite outside the scale range
## (`stat_density()`).
## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).
## Warning: Removed 1 row containing non-finite outside the scale range
## (`stat_density()`).

Customise labels (except for z_labels_chr which currently does not work with this type of plot) and stratify by third categorical variable 
depict(X, x_vars_chr = "phq9_total", x_labels_chr = NA_character_,  y_vars_chr = "k6_total", 
       y_labels_chr = "K6", z_vars_chr = "round", drop_missing_1L_lgl = T, what_1L_chr = "scatterhist")

Contingency table

Balloon plot

Basic balloon plot 
depict(X9, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", 
       what_1L_chr = "balloonplot")
## Ignoring unknown labels:
## • colour : "n"
## • shape : "n"
## • linetype : "n"

Single colour balloon plot 
depict(X9, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", what_1L_chr = "balloonplot",  fill_single_1L_lgl = T)
## Ignoring unknown labels:
## • fill : "n"
## • colour : "n"
## • shape : "n"
## • linetype : "n"

Fized size, labelled balloons 
depict(X9, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", 
       what_1L_chr = "balloonplot", size = 10, show.label=T)
## Ignoring unknown labels:
## • colour : "n"
## • shape : "n"
## • linetype : "n"

Change pallete (from which 3 colours will be selected) 
depict(X9, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", 
       style_1L_chr = "C", type_1L_chr = "viridis", what_1L_chr = "balloonplot", size = 10, show.label=T)
## Ignoring unknown labels:
## • colour : "n"
## • shape : "n"
## • linetype : "n"

Change pallete (to specify a number of colours other than 3) 
depict(X9, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", 
       what_1L_chr = "balloonplot", palette = c("blue",  "red"), size = 10, show.label=T)
## Ignoring unknown labels:
## • colour : "n"
## • shape : "n"
## • linetype : "n"

Facet by up to two additional variables 
depict(X10, x_vars_chr = "d_studying_working", y_vars_chr = "d_sex_birth_s", z_vars_chr = "n", 
       what_1L_chr = "balloonplot", 
       facet.by = c("c_p_diag_s", "c_clinical_staging_s"), 
       ggtheme = ggplot2::theme_bw(), palette = c("blue",  "red"))
## Ignoring unknown labels:
## • colour : "n"
## • shape : "n"
## • linetype : "n"