Bivariate plot

Description

Plots a bivariate graph to visually assess associations. It will be a scatterplot if both variables are numeric and grouped boxplots if one of them is categorical.

plot_bivariate(
  data,
  var,
  area = NULL,
  facet = FALSE,
  free_x_scale = FALSE,
  free_y_scale = FALSE,
  title = NULL,
  var_label = NULL,
  legend = NULL,
  palette = NULL
)

Arguments

  • data: Data frame containing equally spaced (daily, weekly, monthly) covariate or disease case observations for one or multiple locations.
  • var: Character vector of covariate names with length 2 (x, y). One of them can be a factor.
  • area: Character, the name of the variable that identifies the different areal units or any other grouping of interest. If specified, results are grouped by this variable. Defaults to NULL (no grouping).
  • facet: If TRUE, plot each grouping in a different facet.
  • free_x_scale: If TRUE and facet=TRUE, the x-axis scale is free in each facet.
  • free_y_scale: If TRUE and facet=TRUE, the y-axis scale is free in each facet.
  • title: Optional title of the plot.
  • var_label: A 2 character vector with a custom name for the variables.
  • legend: A character vector with a custom name for the legend.
  • palette: GHR, RColorBrewer or colorspace palette (e.g. “Purp”). Single R colors in colors() or hex codes can be used when there is no grouping or facets are used. Use “-” before the palette name (e.g., “-Reds”) to reverse it. Defaults to a dark green when area is NULL or when facet is TRUE, otherwise defaults to the “IDE2” palette.

Returns

A ggplot2 scatterplot or boxplot graph.

Examples

# Load data
data("dengue_MS")

# Scatter (two numeric variables) - No grouping
plot_bivariate(dengue_MS, 
               var = c("pop_density", "tmin"), 
               palette = "#d04a2d")

# Scatter (two numeric variables) - Grouping in the same graph
plot_bivariate(dengue_MS, 
               var = c("pop_density", "tmin"),
               var_label = c("Pop. density", "Min temp."),
               area = "micro_code")

# Scatter  (two numeric variables) - Grouping in facets
plot_bivariate(dengue_MS,
               var = c("pop_density", "tmin"),
               var_label = c("Pop. density", "Min temp."),
               area = "micro_code", facet = TRUE, 
               free_x_scale = TRUE)

# Boxplots (one numeric, one categorical) - No grouping
plot_bivariate(dengue_MS, 
               var = c("pop_density", "biome_name"), 
               var_label = c("Pop. density", "Min temp."),
               palette = "royalblue")

# Boxplots (one numeric, one categorical) - Grouping
plot_bivariate(dengue_MS, 
               var = c("biome_name", "tmin"), 
               area = "meso_code",
               palette = "Accent")