disability-figures.R

# Set-up
library(tidyverse)

# Parameters
point.size <- 6
font.family <- "Aptos Narrow"
font.size <- 10
line.width <- 0.2
line.color <- "black"
text.color <- "black"


# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
#  Figure 1
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

# Reading dataset - disability and difficulties
data.disability <- readxl::read_xlsx("raw_data/Monograph template - tables for graphs.xlsx", sheet = "Disability and difficulties") |>
  mutate(theme = factor(theme, levels = unique(theme), labels = toupper(unique(theme))),
         indicator = factor(indicator, levels = rev(indicator)))

# Themes
data.disability.themes <- data.disability |>
  group_by(theme) |>
  summarise(n = n())

# Creating the bar plot - Disability and difficulties
ggplot(data.disability) +
  geom_col(aes(x = percentage, y = indicator), width = 0.6, fill = "#64C9D8") +
  geom_text(aes(x = percentage + 1, y = indicator, label = paste0(sprintf(percentage, fmt = '%#.1f'), "%")),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0) +
  # Facets
  ggforce::facet_col(facets = vars(theme), scales = "free_y", space = "free") +
  # Y axis text
  geom_text(data = data.disability.themes, aes(x = -42, y = n + 1, label = theme),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0, fontface = "bold") +
  geom_text(aes(x = -42, y = indicator, label = indicator),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0) +
  # Number of people
  geom_rect(data = data.disability.themes, aes(xmin = 101, xmax = 114, ymin = -Inf, ymax = n + 0.5),
            fill = "gray93") + 
  geom_text(data = data.disability |> slice(1), x = 113, y = 2, label =  "People",
            size = font.size/.pt, color = "black", family = font.family, hjust = 1, fontface = "bold") +
  geom_text(data = data.disability, aes(x = 113, y = indicator, label = scales::comma(n)),
            size = font.size/.pt, color = "black", family = font.family, hjust = 1) +
  # Cosmetic lines bellow strips
  geom_segment(data = data.disability.themes, aes(x = -42, xend = 114, y = n + 0.5),
               linewidth = line.width, color = "#64C9D8") +
  # Scales
  scale_x_continuous(breaks = seq(0,100,20), expand = expansion(add = 0)) +
  coord_cartesian(clip = "off", xlim = c(0,100)) + 
  theme(# Axis: 
    axis.ticks.y = element_blank(),
    axis.ticks.x = element_line(color = line.color, linewidth = line.width),
    axis.line.x = element_line(color = line.color, linewidth = line.width),
    axis.text.x = element_text(size = font.size, family = font.family, color = text.color),
    axis.text.y = element_blank(),
    axis.title.y = element_blank(),
    axis.title.x = element_text(size = font.size, family = font.family, color = text.color),
    # Legend: 
    legend.text = element_text(size = font.size, family = font.family, color = text.color),
    legend.title = element_blank(),
    legend.position = "bottom",
    # Strip: 
    strip.text = element_blank(),
    strip.background = element_blank(),
    # Panel:
    panel.background = element_blank(),
    panel.grid = element_blank(),
    panel.spacing.y = unit(20, "pt"),
    # Plot
    plot.margin = margin(10,40,5,118)) +
  labs(x = "Percentage of the population aged 5 years or older") 

# Saving the figure for report
ggsave("output/figure_1.pdf", height = 4, width = 6, device = cairo_pdf)


# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
#  Figure 2
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-


# Reading dataset - life outcomes
data.outcomes <- readxl::read_xlsx("raw_data/Monograph template - tables for graphs.xlsx", sheet = "Life outcomes") |>
  mutate(theme = factor(theme, levels = unique(theme), labels = toupper(unique(theme))),
         indicator = factor(indicator, levels = rev(indicator)))

# Long format
data.outcomes.long <- data.outcomes |>
  pivot_longer(4:5, names_to = "disability", values_to = "percentage")

# Themes
data.outcomes.themes <- data.outcomes |>
  group_by(theme) |>
  summarise(n = n())

# Creating the equiplot - Life outcomes

ggplot(data.outcomes.long) +
  # Equiplot lines and points
  geom_line(aes(x = percentage, y = indicator, group = indicator),
            linewidth = point.size, color = "gray93") +
  geom_point(data = data.outcomes, aes(x = nodisability, y = indicator, color = "Persons without disability"),
             stroke = 0, size = point.size) +
  geom_text(data = data.outcomes, aes(x = nodisability, y = indicator, label = round(nodisability)),
            size = 9/.pt, color = "white", family = font.family) +
  geom_point(data = data.outcomes, aes(x = disability, y = indicator, color = "Persons with disability"),
             stroke = 0, size = point.size) +
  geom_text(data = data.outcomes, aes(x = disability, y = indicator, label = round(disability)),
            size = 9/.pt, color = "black", family = font.family) +
  # Y axis text
  geom_text(data = data.outcomes.themes, aes(x = -42, y = n + 1, label = theme),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0, fontface = "bold") +
  geom_text(data = data.outcomes, aes(x = -42, y = indicator, label = indicator),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0) +
  # Age
  geom_rect(data = data.outcomes.themes, aes(xmin = 101, xmax = 111, ymin = -Inf, ymax = n + 0.5),
            fill = "gray93") + 
  geom_text(data = data.outcomes |> slice(1), x = 106, y = 9, label =  "Age",
            size = font.size/.pt, color = "black", family = font.family, hjust = 0.5, fontface = "bold") +
  geom_text(data = data.outcomes, aes(x = 106, y = indicator, label = age),
            size = font.size/.pt, color = "black", family = font.family, hjust = 0.5) +
  # Cosmetic lines bellow strips
  geom_segment(data = data.outcomes.themes, aes(x = -42, xend = 111, y = n + 0.5),
               linewidth = line.width, color = "#64C9D8") +
  # Facets
  ggforce::facet_col(facets = vars(theme), scales = "free_y", space = "free") +
  # Scales
  scale_x_continuous(breaks = seq(0,100,20), expand = expansion(add = 0)) +
  scale_color_manual(values = c("Persons with disability" = "#64C9D8", 
                                "Persons without disability" = "#004EC5")) + 
  coord_cartesian(clip = "off", xlim = c(0,100)) + 
  theme(# Axis: 
    axis.ticks.y = element_blank(),
    axis.ticks.x = element_line(color = line.color, linewidth = line.width),
    axis.line.x = element_line(color = line.color, linewidth = line.width),
    axis.text.x = element_text(size = font.size, family = font.family, color = text.color),
    axis.text.y = element_blank(),
    axis.title.y = element_blank(),
    axis.title.x = element_text(size = font.size, family = font.family, color = text.color),
    # Legend: 
    legend.text = element_text(size = font.size, family = font.family, color = text.color),
    legend.title = element_blank(),
    legend.position = "bottom",
    # Strip: 
    strip.text = element_blank(),
    strip.background = element_blank(),
    # Panel:
    panel.background = element_blank(),
    panel.grid = element_blank(),
    panel.spacing.y = unit(20, "pt"),
    # Plot
    plot.margin = margin(10,33,5,120)) +
  labs(x = "Percentage of the population")

# Saving the figure for report
ggsave("output/figure_2.pdf", height = 9, width = 6, device = cairo_pdf)


# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
#  Figure 3
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

# Reading dataset - disability and age
data.age <- readxl::read_xlsx("raw_data/Monograph template - tables for graphs.xlsx", sheet = "Disability and age") |>
  mutate(age = factor(age, levels = unique(age))) |>
  pivot_longer(2:4, names_to = "group", values_to = "percentage") |>
  mutate(group = factor(group, levels = c("total", "female", "male"),
                        labels = toupper(c("Total population", "Women", "Men"))))

# Creating the bar plot - disability and age
ggplot(data.age |> filter(age != "National")) +
  aes(x = age, y = percentage, fill = group) +
  geom_col(width = 0.5, show.legend = FALSE) +
  geom_hline(data = data.age |> filter(age == "National"),
             aes(yintercept = percentage, color = group),
             linewidth = line.width*.stroke/.pt, linetype = "dashed", show.legend = FALSE) +
  geom_text(data = data.age |> filter(age == "National"),
            aes(y = percentage + 2, x = 0.7,
                label = paste0("National prevalence: ", sprintf(percentage, fmt = '%#.1f'), "%")),
            size = font.size/.pt, color = "black", family = font.family, vjust = 0, hjust = 0) +
  geom_text(aes(y = percentage + 1, label = sprintf(percentage, fmt = '%#.1f')),
            size = 9/.pt, color = "black", family = font.family, vjust = 0) +
  # x Axis
  geom_segment(aes(x = -Inf, xend = Inf, y = 0),
               linewidth = line.width, color = line.color) +
  # Facets
  ggforce::facet_col(facets = vars(group)) +
  # Scales
  scale_fill_manual(values = c("gray50", "#64C9D8", "#004EC5")) + 
  scale_color_manual(values = c("gray50", "#64C9D8", "#004EC5")) + 
  scale_y_continuous(breaks = seq(0,60,15), expand = expansion(add = 0), limits = c(0,60)) +
  scale_x_discrete(expand = expansion(add = c(0.4, 0.25))) +
  coord_cartesian(clip = "off") +
  # Theme
  theme(# Axis: 
    axis.ticks = element_line(color = line.color, linewidth = line.width),
    axis.line = element_line(color = line.color, linewidth = line.width),
    axis.text = element_text(size = font.size, family = font.family, color = text.color),
    axis.title = element_text(size = font.size, family = font.family, color = text.color),
    # Strip: 
    strip.text = element_text(size = font.size, family = font.family, color = text.color, face = "bold"),
    strip.background = element_blank(),
    # Panel:
    panel.background = element_blank(),
    panel.grid = element_blank(),
    panel.spacing.y = unit(15, "pt")) +
  labs(x = "Age group (years)",
       y = "Prevalence of disability")

# Saving the figure for report
ggsave("output/figure_3.pdf", height = 6, width = 6, device = cairo_pdf)


# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
#  Figure 4
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

# Reading dataset - disability and age
data.school <- readxl::read_xlsx("raw_data/Monograph template - tables for graphs.xlsx", sheet = "School attendance")

gap.y <- (sum(data.school$nodisability[data.school$age %in% 11:12]) + sum(data.school$disability[data.school$age %in% 11:12]))/4

# Creating the bar plot - disability and school attendance
ggplot(data.school) +
  aes(x = age) +
  geom_ribbon(aes(ymin = disability, ymax = nodisability),
              fill = "gray80", alpha = 0.5) +
  # Attendance gap
  annotate("text", x = 11.5, y = gap.y,
           label = "ATTENDANCE GAP", hjust = 0.5, fontface = "bold",
           size = font.size/.pt, color = "black", family = font.family) +
  # Lines
  geom_line(aes(y = nodisability), color = "#004EC5", linewidth = 0.6) +
  annotate("text", x = 11.5, y = 105,
           label = "Children without disabilities", hjust = 0.5,
           size = font.size/.pt, color = "#004EC5", family = font.family) +
  geom_line(aes(y = disability), color = "#64C9D8", linewidth = 0.6) +
  annotate("text", x = 11.5, y = 55,
           label = "Children with disabilities", hjust = 0.5,
           size = font.size/.pt, color = "#00818F", family = font.family) +
  # Scales
  scale_x_continuous(breaks = 5:18, expand = expansion(add = 0)) +
  scale_y_continuous(breaks = seq(0,100,20), expand = expansion(add = 0)) +
  coord_cartesian(clip = "off", xlim = c(5,18), ylim = c(0,100)) +
  # Theme
  theme(# Axis: 
    axis.ticks.x = element_line(color = line.color, linewidth = line.width),
    axis.ticks.y = element_blank(),
    axis.line.x = element_line(color = line.color, linewidth = line.width),
    axis.line.y = element_blank(),
    axis.text = element_text(size = font.size, family = font.family, color = text.color),
    axis.title = element_text(size = font.size, family = font.family, color = text.color),
    # Legend: 
    legend.text = element_text(size = font.size, family = font.family, color = text.color),
    legend.title = element_blank(),
    legend.position = "bottom",
    # Panel:
    panel.background = element_blank(),
    panel.grid.major.x = element_blank(),
    panel.grid.major.y = element_line(color = "gray", linewidth = line.width),
    panel.grid.minor = element_blank(),
    # Plot
    plot.margin = margin(15,5,5,5)) +
  labs(x = "Age (years)", y = "Percentage of children currently attending school") 

# Saving the figure for report
ggsave("output/figure_4.pdf", height = 3, width = 6, device = cairo_pdf)