README.md

In jorvlan/openvis: Create visualizations based on best open science practices

This repository has moved to a different location, please see:

https://github.com/jorvlan/raincloudplots

This repository serves as the R package for the revisited version of: Raincloud plots: a multi-platform tool for robust data visualization (to be published).

An example figure

Package demonstration

• Tested on macOS with R version >= 4.0.1
• Tested on Windows with R version >= 4.0.1
• If installing does not work, please first see 'debug.R'
• If you want to plug and play to create rainclouds manually with geom's, and not with the functions from in this package, please see:
  • raw_tutorial
  • https://github.com/jorvlan/open-visualizations
  • https://github.com/RainCloudPlots/RainCloudPlots
• Please note that this package is not officially published yet, awaiting manuscript revisioning.

Installation

if (!require(devtools)) {
    install.packages('devtools')
}
devtools::install_github('jorvlan/openvis')
library("openvis")

Raincloud 1 x 1 comparison

Step 1: Initialize the data-format

df_1x1 <- data_1x1(
  array_1 = iris$Sepal.Length[1:50],
  array_2 = iris$Sepal.Length[51:100],
  jit_distance = .09,
  jit_seed = 321)

> head(df_1x1)
  y_axis x_axis id       jit
1    5.1      1  1 1.0820609
2    4.9      1  2 1.0787114
3    4.7      1  3 0.9528797
4    4.6      1  4 0.9559133
5    5.0      1  5 0.9802922
6    5.4      1  6 0.9714124
> tail(df_1x1)
    y_axis x_axis id      jit
95     5.6      2 45 2.059387
96     5.7      2 46 2.004848
97     5.7      2 47 2.066980
98     6.2      2 48 2.074479
99     5.1      2 49 1.939248
100    5.7      2 50 1.999004

Step 2: Create a vertical or horizontal 1 x 1 Raincloud

figure_1x1_v <- raincloud_1x1(df_1x1, ort = 'v') +

  scale_x_continuous(breaks=c(1,2), labels=c("Group1", "Group2"), limits=c(0, 3)) +
  xlab("Groups") + 
  ylab("Score") +
  theme_classic()

figure_1x1_v

figure_1x1_h <- raincloud_1x1(df_1x1, ort = 'h') +

  scale_x_continuous(breaks=c(1,2), labels=c("Group1", "Group2"), limits=c(0, 3)) +
  xlab("Groups") + 
  ylab("Score") +
  theme_classic()

figure_1x1_h

Raincloud repeated measures

figure_1x1_rm <- raincloud_1x1_repmes(df_1x1, align_clouds = FALSE) +

  scale_x_continuous(breaks=c(1,2), labels=c("Before", "After"), limits=c(0, 3)) +
  xlab("States") + 
  ylab("Score") +
  theme_classic()

figure_1x1_rm

figure_1x1_rm_2.0 <- raincloud_1x1_repmes(df_1x1, align_clouds = TRUE) +

  scale_x_continuous(breaks=c(1,2), labels=c("Before", "After"), limits=c(0, 3)) +
  xlab("States") + 
  ylab("Value") +
  theme_classic()

figure_1x1_rm_2.0

Raincloud 2 x 2 repeated measures

Step 1: Initialize the data-format

df_2x2_1.0 <- data_2x2(
  array_1 = iris$Sepal.Length[1:50],
  array_2 = iris$Sepal.Length[51:100],
  array_3 = iris$Sepal.Length[101:150],
  array_4 = iris$Sepal.Length[81:130],
  label_1 = 'congruent',
  label_2 = 'incongruent',
  jit_distance = .05,
  jit_seed = 321,
  spread_x_ticks = TRUE) # FALSE if 2 x-ticks

> head(df_2x2_1.0)
  y_axis x_axis id     group       jit
1    5.1      1  1 congruent 1.0455894
2    4.9      1  2 congruent 1.0437286
3    4.7      1  3 congruent 0.9738220
4    4.6      1  4 congruent 0.9755074
5    5.0      1  5 congruent 0.9890512
6    5.4      1  6 congruent 0.9841180
> tail(df_2x2_1.0)
    y_axis x_axis id       group      jit
195    6.7      4 45 incongruent 4.025752
196    7.2      4 46 incongruent 3.980672
197    6.2      4 47 incongruent 4.000718
198    6.1      4 48 incongruent 4.001973
199    6.4      4 49 incongruent 3.972786
200    7.2      4 50 incongruent 4.042541

Step 2: Create a raincloud plot with 4 x-ticks or with 2 x-ticks

figure_2x2_1.0 <- raincloud_2x2_repmes(df_2x2_1.0, spread_x_ticks = TRUE) +

  scale_x_continuous(breaks=c(1,2,3,4),
                     labels=c("low-congr", "high-congr", "low-incongr", "high-incongr"),
                     limits=c(0, 5)) +
  xlab("Conditions") +
  ylab("Score") +
  theme_classic()

figure_2x2_1.0

df_2x2_2.0 <- data_2x2(
  array_1 = iris$Sepal.Length[1:50],
  array_2 = iris$Sepal.Length[51:100],
  array_3 = iris$Sepal.Length[101:150],
  array_4 = iris$Sepal.Length[81:130],
  label_1 = 'congruent',
  label_2 = 'incongruent',
  jit_distance = .05,
  jit_seed = 321,
  spread_x_ticks = FALSE)
 ```
```r
> head(df_2x2_2.0)
  y_axis x_axis id     group       jit
1    5.1      1  1 congruent 1.0455894
2    4.9      1  2 congruent 1.0437286
3    4.7      1  3 congruent 0.9738220
4    4.6      1  4 congruent 0.9755074
5    5.0      1  5 congruent 0.9890512
6    5.4      1  6 congruent 0.9841180
> tail(df_2x2_2.0)
    y_axis x_axis id       group      jit
195    6.7   2.01 45 incongruent 2.035752
196    7.2   2.01 46 incongruent 1.990672
197    6.2   2.01 47 incongruent 2.010718
198    6.1   2.01 48 incongruent 2.011973
199    6.4   2.01 49 incongruent 1.982786
200    7.2   2.01 50 incongruent 2.052541

figure_2x2_2.0 <- raincloud_2x2_repmes(df_2x2_2.0, spread_x_ticks = FALSE) +

  scale_x_continuous(breaks=c(1,2), labels=c("Before", "After"), limits=c(0, 3)) +
  xlab("States") +
  ylab("Score") +
  theme_classic()

figure_2x2_2.0

Raincloud 2 x 3 (repeated measures)

Step 1: Initialize the data-format

df_2x3 <- data_2x2(
  array_1 = iris$Sepal.Length[1:50],
  array_2 = iris$Sepal.Length[51:100],
  array_3 = iris$Sepal.Length[101:150],
  array_4 = iris$Sepal.Length[81:130],
  array_5 = iris$Sepal.Length[21:70],
  array_6 = iris$Sepal.Length[41:90],
  label_1 = 'Drug',
  label_2 = 'Placebo',
  jit_distance = .07,
  jit_seed = 321)

> head(df_2x3)
  y_axis x_axis id group       jit
1    5.1      1  1  Drug 1.0638251
2    4.9      1  2  Drug 1.0612200
3    4.7      1  3  Drug 0.9633509
4    4.6      1  4  Drug 0.9657103
5    5.0      1  5  Drug 0.9846717
6    5.4      1  6  Drug 0.9777652
> tail(df_2x3)
    y_axis x_axis id   group      jit
295    5.4   3.01 45 Placebo 3.073854
296    6.0   3.01 46 Placebo 3.062774
297    6.7   3.01 47 Placebo 3.077949
298    6.3   3.01 48 Placebo 3.003069
299    5.6   3.01 49 Placebo 2.965916
300    5.5   3.01 50 Placebo 2.950490

Step 2: Create a vertical or horizontal 2 x 3 Raincloud

figure_2x3_v <- raincloud_2x3_repmes(df_2x3, ort = 'v') +

  scale_x_continuous(breaks=c(1,2,3),
                     labels=c("Time-1", "Time-2", "Time-3"),
                     limits=c(0, 4)) +
  xlab("States") +
  ylab("Score") +
  theme_classic()

figure_2x3_v

figure_2x3_h <- raincloud_2x3_repmes(df_2x3, ort = 'h') +

  scale_x_continuous(breaks=c(1,2,3),
                     labels=c("1", "2", "3"),
                     limits=c(0, 4)) +
  xlab("States") +
  ylab("Score") +
  theme_classic()

figure_2x3_h

To-do package

• set CRAN mirror automatically
• reduce dependencies
• change package name
• user ID error -> https://stackoverflow.com/questions/30599326/warning-message-during-building-an-r-package-invalid-uid-value-replaced-by-that
• try remotes
• add ggplot extensions in functions (color options) to function names -> e.g., fill= col= in de tutorial
• When opening/installing package, refer to tutorial (if possible).
• When done, share with some to test installation.

To-do GitHub

• Explicitly mention for which designs this package works in README.md
• Change GitHub Repository name

To-do paper

• Rogier: start R part with script / package.
• Jordy: Write with which examples one can make RM.
• Jordy: Write down the added value of my part, e.g., more robust, more transparent open science, repeated measures extension, method already adopted in 15 papers (see list on open-visualizations GitHub).
• Jordy: add the two script examples and package lines parts to the rainclouds_tutorial_r.Rmd
• Mention that if you want to make Rainclouds with more groups, more timepoints, more modular -> look at the code RainCloudPlots and open-visualizations.
• Mention 2 example plots from the package (1x1, 2x2) with additional comments on how to install the package.

jorvlan/openvis documentation built on Dec. 6, 2020, 3:06 a.m.