data visualization R code.Rmd

---
title: "Data Visualization in R"
author: "Annalise LaPlume"
date: "`r format(Sys.time(), '%B %d, %Y')`"
output:
  html_document:
    highlight: textmate
    theme: paper
    toc: yes
    toc_depth: 3
    toc_float: true
  pdf_document:
    toc: yes
editor_options:
  chunk_output_type: console
---

```{r setup, include=FALSE, error=TRUE}
## load packages
if(!require("car"))install.packages("car") # Basic functions
if(!require("plyr"))install.packages("plyr") 
if(!require("base"))install.packages("base")
if(!require("hablar"))install.packages("hablar")
if(!require("tidyverse"))install.packages("tidyverse")
if(!require("psych"))install.packages("psych") # Calculate descriptive statistics
if(!require("ggplot2"))install.packages("ggplot2") # Plot results
if(!require("raincloudplots"))install.packages("raincloudplots") # Raincloud plots
if(!require("ggpubr"))install.packages("ggpubr") # Plot functions (e.g., correlations)


source("R_rainclouds.R") # Raincloud plots scripts
source("summarySE.R") # Raincloud plots calculations

## Figure settings
# width and height variables for saved plots
w = 6
h = 3 

```

```{r import data, include=FALSE, error=TRUE}
## import dataset
ds <- read.csv("C:/Users/.../sample_data.csv"",                     # Enter location of datafile here 
    header = TRUE, sep = ",")

# View sample of data
head(ds)

# Check format of variables
str(ds)

# Reformat variables 
ds <- ds %>% 
  convert(num(score, age),                                          # Specify numberic variables
          fct(id, group, sex, age_group))                           # Specify factors/group variables

# Set baseline (reference) level 
levels(ds$group)                                                    # View levels of group
ds$group = relevel(ds$group, ref="placebo")                         # Reorder levels to placebo as reference
levels(ds$group)
```

# Descriptive statistics

## n
Number of individuals per group
```{r group n, include=TRUE, echo=FALSE}
table(ds$group)
```
Balanced number in each group

## Background variables
Sex
```{r sex n, include=TRUE, echo=FALSE}
table(ds$sex)
```

Age group
```{r age group n, include=TRUE, echo=FALSE}
table(ds$age_group)
```

Age
```{r age, echo=FALSE}
plyr::ddply(ds, ~group, summarise, 
      mean=round(mean(age), 1), sd=round(sd(age), 1),
      min=min(age), max=max(age))
```

## Dependent variables
Test score (Assume lower score=worse, and higher score=better)
```{r score, echo=FALSE}
plyr::ddply(ds, ~group, summarise, 
      mean=round(mean(score), 1), sd=round(sd(score), 1),
      min=min(score), max=max(score))
```

# Distribution plots

## Histogram
```{r histogram, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=score,                                                 # Variable (x)
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_histogram(                                                   # Add histogram
                 fill="grey", colour="black",                       # Format bin colour (fill) and outline (colour)
                 position = "identity",                             # Statistic to present
                 bins = 10,                                         # Number of bins/bars (can also specify binwidth)
                 alpha=0.4) +                                       # Bar transparency
    labs(x="Score", y="Frequency",                                  # Axis labels
         title="Histogram") +                                       # Plot title
       theme_classic() +                                            # Remove background grid
         theme(legend.position = "none",                            # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre  

# Alternate way to draw a histogram
hist(ds$score,                                                      # Variable
      xlab="Score", ylab="Frequency",                               # Axis labels 
      main="Histogram",                                             # Plot title   
     breaks=10)                                                     # Number of bins/bars
```

Different bar widths/number
```{r histogram bar widths, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age,                                                   # Variable (x)
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_histogram(                                                   # Add histogram
                 fill="grey", colour="black",                       # Format bin colour (fill) and outline (colour)
                 position = "identity",                             # Statistic to present
                 bins = 10,                                         # Number of bins/bars (can also specify binwidth)
                 alpha=0.4) +                                       # Bar transparency
    labs(x="Age", y="Frequency",                                    # Axis labels
         title="Histogram (10 bins)") +                             # Plot title
       theme_classic() +                                            # Remove background grid
         theme(legend.position = "none",                            # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre  


ggplot(data=ds,                                                     # Data
       aes(x=age,                                                   # Variable (x)
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_histogram(                                                   # Add histogram
                 fill="grey", colour="black",                       # Format bin colour (fill) and outline (colour)
                 position = "identity",                             # Statistic to present
                 bins = 7,                                          # Number of bins/bars (can also specify binwidth)
                 alpha=0.4) +                                       # Bar transparency
    labs(x="Age", y="Frequency",                                    # Axis labels
         title="Histogram (7 bins)") +                              # Plot title
       theme_classic() +                                            # Remove background grid
         theme(legend.position = "none",                            # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre  
```

## Dotplot
```{r dotplot, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=group, y = score,                                      # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
    labs(x="Group", y="Score",                                      # Axis labels
         title="Dotplot") +                                         # Plot title
       theme_classic() +                                            # Remove background grid
         theme(legend.position = "none",                            # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre  
```

## Boxplot
```{r boxplot, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=group, y = score,                                      # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
   labs(x="Group", y="Score",                                       # Axis labels
        title="Boxplot") +                                          # Plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size                    
              plot.title=element_text(size=14, face="bold",         # Axis font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre
          geom_boxplot(notch=FALSE,                                 # Add boxplot outline
                       alpha=0, width=0.3)                          # Format boxplot


```

Dotplot with superimposed boxplot
```{r boxplot dotplot, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=group, y = score,                                      # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
   labs(x="Group", y="Score",                                       # Axis labels
        title="Boxplot-dotplot") +                                          # Plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size                    
              plot.title=element_text(size=14, face="bold",         # Axis font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre
          geom_boxplot(notch=FALSE,                                 # Add boxplot outline
                       alpha=0, colour="grey46", width=0.3)         # Format boxplot
```

## Violin plot
```{r violin, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=group, y = score,                                      # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
    labs(x="Group", y="Score", title="Violin plot") +               # Axis labels and plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size
              plot.title=element_text(size=14, face="bold",         # Title font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre     
        geom_violin (alpha=0, width=0.3)                            # Add violin plot
         
```

Dotplot with superimposed violin plot
```{r violin dotplot, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=group, y = score,                                      # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
    labs(x="Group", y="Score", title="Violin-dotplot") +               # Axis labels and plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size
              plot.title=element_text(size=14, face="bold",         # Title font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre     
        geom_violin (                                               # Add violin plot outline
                      alpha=0, colour="grey46", width=0.3) +        # Format violin plot
          stat_summary(fun.data = "mean_sdl",                       # Add mean and error bars     
                       fun.args = list(mult = 1), 
                       geom = "pointrange",
                       color = "black", size=0.2)                    
```

## Boxplot with many groups/categories
```{r boxplot groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age_group, y = score,                                  # Variables (x, y)
           colour=age_group, fill=age_group,                        # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
   labs(x="Age Group", y="Score",                                   # Axis labels
        title="Boxplot with many groups") +                         # Plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size
              plot.title=element_text(size=14, face="bold",         # Title font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre     
          geom_boxplot(notch=FALSE,                                 # Add boxplot outline
                       alpha=0)                                     # Format boxplot
```

## Violin plot with many groups/categories
```{r violin groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age_group, y = score,                                  # Variables (x, y)
           colour=age_group,                                        # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
    labs(x="Group", y="Score",                                      # Axis labels and
         title="Violin plot with many groups") +                    # Plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre     
          geom_violin (alpha=0) +                                   # Format violin plot
            stat_summary(fun.data = "mean_sdl",                     # Add mean and error bars     
                         fun.args = list(mult = 1), 
                         geom = "pointrange",
                         color = "grey", size=0.2) 
```

# x-y plots
## Scatterplot
```{r scatterplot, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score, colour=age,                            # Variables (x, y)
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
    labs(x="Age", y="Score", title="Scatterplot") +                 # Axis labels and plot title
      theme_classic() +                                             # Format background
        theme(legend.position = "none",                             # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre     
```


## Scatterplot with groups
Two groups
```{r scatterplot 2 groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score,                                        # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
  labs(x="Age", y="Score", colour="Group",                          # Axis labels
       title="Scatterplot with two groups") +                       # Plot title
      theme_classic() +                                             # Format background
    theme(axis.text=element_text(size=12),                          # Axis elements font size
          axis.title=element_text(size=12),                         # Axis font size
           plot.title=element_text(size=14, face="bold",            # Title font size and bolding
                                    hjust = 0.5))                   # Title reposition to centre     
```

Many groups
```{r scatterplot many groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score, colour=age_group,                      # Variables (x, y)
           na.rm= TRUE)) +                                          # Remove missing values      
 geom_point(size = 1.2,                                             # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
  labs(x="Age", y="Score", colour="Age Group",                      # Axis labels
       title="Scatterplot with many groups") +                      # Plot title
      theme_classic() +                                             # Format background
  theme(axis.text=element_text(size=12),                            # Axis elements font size
        axis.title=element_text(size=12),                           # Axis font size
        plot.title=element_text(size=14, face="bold",               # Title font size and bolding
                                hjust = 0.5))                       # Title reposition to centre     
```

## Line graph
Linear regression, line of best fit
```{r line graph, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score, colour=age,                            # Variables (x, y)
           na.rm= TRUE)) +                                          # Remove missing values      
 geom_point(size = 1.2,                                             # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
  labs(x="Age", y="Score",                                          # Axis labels
       title="Line graph") +                                        # Plot title
          theme_classic() +                                         # Format background
            theme(legend.position = "none",                         # Remove legend
                  axis.text=element_text(size=12),                  # Axis elements font size
                  axis.title=element_text(size=12),                 # Axis font size
                  plot.title=element_text(size=14, face="bold",     # Title font size and bolding
                                          hjust = 0.5)) +           # Title reposition to centre     
                    geom_smooth(method = "lm") +                    # straight line graph, 95% CI shaded in grey
                      stat_cor(method = "pearson")                  # Pearson correlation
```

Add effect size (R-squared)
```{r line graph R2, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score, colour=age      ,                      # Variables (x, y)
           na.rm= TRUE)) +                                          # Remove missing values      
 geom_point(size = 1.2,                                             # Data points (size)
            alpha = .4,                                             # Data points (transparency)
            position = position_jitter(width = 0.1)) +              # Data points (random variation)
  labs(x="Age", y="Score",                                          # Axis labels
       title="Line graph") +                                        # Plot title
    theme_classic() +                                               # Format background
      theme(legend.position = "none",                               # Remove legend
            axis.text=element_text(size=12),                        # Axis elements font size
            axis.title=element_text(size=12),                       # Axis font size
            plot.title=element_text(size=14, face="bold",           # Title font size and bolding
                                    hjust = 0.5)) +                 # Title reposition to centre     
        geom_smooth(method = "lm") +                                # straight line graph, 95% CI shaded in grey
          stat_cor(method = "pearson",                              # Pearson correlation
            aes(label = paste(..rr.label..,                         # R-squared
                              ..p.label.., sep = "~`,`~")))         # p-value 
```

## Line graph with two groups

Two groups, both lines on the same graph
```{r line graph 2 groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score,                                        # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
geom_point(size = 1.2,                                              # Data points (size)
           alpha = .4,                                              # Data points (transparency)
           position = position_jitter(width = 0.1)) +               # Data points (random variation)
  labs(x="Age", y="Score", colour="Group",                          # Axis labels
       title="Line graph by Group") +                               # Plot title
    theme_classic() +                                               # Format background
      geom_smooth(method = "lm") +                                  # straight line graph, 95% CI shaded in grey
      stat_cor(method = "pearson",                                  # Pearson correlation
              aes(label = paste(..rr.label..,                       # R-squared  
                                ..p.label.., sep = "~`,`~"))) +     # p-value 
        theme(legend.position = "none",                             # Remove legend
              axis.text=element_text(size=12),                      # Axis elements font size
              axis.title=element_text(size=12),                     # Axis font size
             plot.title=element_text(size=14,                       # Title font size 
                                     face="bold",                   # Title bolding
                                     hjust = 0.5))                  # Title reposition to centre   
```

Two groups, side-by-side
```{r line graph many groups, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score,                                        # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
 geom_point(size = 1.2,                                             # Data points (size)
            alpha = .4,                                             # Data points (transparency)
            position = position_jitter(width = 0.1)) +              # Data points (random variation)
  labs(x="Age", y="Score", colour="Group",                          # Axis labels
       title="Line graph by Group") +                               # Plot title
          theme_classic() +                                         # Format background
            geom_smooth(method = "lm") +                            # Straight line graph, 95% CI shaded in grey
              stat_cor(method = "pearson",                          # Pearson correlation
                       aes(label = paste(..rr.label..,              # R-squared
                           ..p.label.., sep = "~`,`~"))) +          # p-value 
                facet_wrap(. ~ group) +                             # Facet into separate graphs
                  theme(legend.position = "none",                   # Remove legend
                        axis.text=element_text(size=12),            # Axis elements font size
                        axis.title=element_text(size=12),           # Axis font size
                        plot.title=element_text(size=14,            # Title font size 
                                                face="bold",        # Title bolding
                                                hjust = 0.5),       # Title reposition to centre
                         strip.text = element_text(size = 14))      # Text size of panel
```


## Smooth line graphs 
Locally weighted regression scatterplot smoothing (loess)
```{r loess, echo=FALSE}
ggplot(data=ds,                                                     # Data
       aes(x=age, y = score, colour=age,                            # Variables (x, y)
           na.rm= TRUE)) +                                          # Remove missing values      
 geom_point(size = 1.2,                                             # Data points (size)
            alpha = .4,                                             # Data points (transparency)
            position = position_jitter(width = 0.1)) +              # Data points (random variation)
  labs(x="Age", y="Score",                                          # Axis labels
       title="Smooth line graph") +                                 # Plot title
    theme_classic() +                                               # Format background
      theme(legend.position = "none",                               # Remove legend
            axis.text=element_text(size=12),                        # Axis elements font size
            axis.title=element_text(size=12),                       # Axis font size
            plot.title=element_text(size=14,                        # Title font size 
                                    face="bold",                    # Title bolding
                                    hjust = 0.5)) +                 # Title reposition to centre      
              geom_smooth(method = "loess",                         # straight line graph, 95% CI shaded in grey
                          span=1)                                   # control amount of smoothing        
```


## Slope graph 

```{r import repeated measures data, include=FALSE, error=TRUE}
## import dataset
ds_rep <- read.csv("C:/Users/.../rep_measures_data.csv", 
    header = TRUE, sep = ",")

# View sample of data
head(ds_rep)                                                        # Arranged in long format

# Check format of variables
str(ds_rep)

# Reformat variables 
ds_rep <- ds_rep %>% 
  convert(num(score),                                               # Specify numberic variables
          fct(id, group, session))                                  # Specify factors/group variables

# Set baseline (reference) level 
levels(ds_rep$group)                                                # View levels of group
ds_rep$group = relevel(ds_rep$group, ref="placebo")                 # Reorder levels to placebo as reference
levels(ds_rep$group)                                                # Check levels of group

levels(ds_rep$session)                                              # View levels of session
ds_rep$session = relevel(ds_rep$session, ref="initial")             # Reorder levels to initial as reference
levels(ds_rep$session)                                              # Check levels of session
```

View longitudinal dotplot
```{r dotplot repeated measures, echo=FALSE}
ggplot(data=ds_rep,                                                 # Data
       aes(x=session, y = score,                                    # Variables (x, y)
           colour=group,                                            # Grouping variable
           na.rm= TRUE)) +                                          # Remove missing values      
  geom_point(size = 1.2,                                            # Data points (size)
             alpha = .4,                                            # Data points (transparency)
             position = position_jitter(width = 0.1)) +             # Data points (random variation)
    labs(x="Session", y="Score",                                    # Axis labels
         title="Dotplot for repeated measures") +                   # Plot title
       theme_classic() +                                            # Remove background grid
         theme(legend.position = "none",                            # Remove legend
               axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5))                # Title reposition to centre  
```


```{r slope graph, echo=FALSE}
ggplot(data=ds_rep,                                                 # Data
       aes(x=session, y = score,                                    # Variables (x, y)
           colour=group,                                            # Grouping variable
           group=id,                                                # Link participant scores to each other
           na.rm= TRUE)) +                                          # Remove missing values      
    labs(x="Session", y="Score",                                    # Axis labels
         title="Slope graph for repeated measures") +               # Plot title
       theme_classic() +                                            # Remove background grid
         theme(axis.text=element_text(size=12),                     # Axis elements font size
               axis.title=element_text(size=12),                    # Axis elements font size
               plot.title=element_text(size=14, face="bold",        # Title font size and bolding
                                       hjust = 0.5)) +              # Title reposition to centre  
            geom_line(size = 1.2, alpha = .4,                       # Add lines
                      position = position_jitter(width = 0.2, 
                                                 height=0.5))
```