ggplot具有不同数量x轴类别但条形宽度恒定的堆叠条形图

Question

I have generating multiple stacked bar charts for a report. Each chart represents a different region and may have different number of x-axis categories. For a consistent visual between different charts I wish ensure that the width of the bars are constant between separate charts. I currently haven't been able to do this as charts with fewer x-axis categories have wider bars. It seems a combination of dodge and width might work for standard bar charts, but this does not work for stacked bars. Note that there are 100s of possible x-axis categories, though only 3 to 6 (max 6) are actually shown on each chart. Does anyone know of a means of specifying a constant bar width which could then be applied to all charts?

One thought I had for regions with less than 6 categories, was to add additional empty categories to the right end of the chart (bringing total to 6) and then somehow set the plot panel width so as to cut these last empty placeholders off. I'm not sure how to implement the cutting off of the placeholders.

Here is a simple example of the basic problem.

df1 <- data.frame(variable_name = c(rep('var1', 3),rep('var2',4), rep('var3', 3), rep('var4', 4)),
             var_value = c(1,4,2,4,3,6,2,1,6,3,1,3,2,4),
             value_type = c('A','B','C','A','B','C','D','B','C','D','B','D','E','F'))

df2 <- data.frame( variable_name = c(rep('var6',4), rep('var4', 4)),
             var_value = c(4,3,6,2,1,3,2,4),
             value_type = c('A','B','C','D','B','D','E','F'))

ggplot() + 
    geom_col(data = df1,
    aes(x = variable_name, y =var_value,fill = value_type), 
    width = 0.8)

ggplot() + 
    geom_col(data = df2,
    aes(x = variable_name, y =var_value,fill = value_type), 
    width = 0.8)

Answer 1

Ultimately bar width will be determined by the width at which you save the plot, but if you want to scale within the plot, just multiply width by the number of x values divided by the maximum possible:

library(ggplot2)

df1 <- data.frame(variable_name = c(rep('var1', 3), rep('var2', 4), rep('var3', 3), rep('var4', 4)),
                  var_value = c(1,4,2,4,3,6,2,1,6,3,1,3,2,4),
                  value_type = c('A','B','C','A','B','C','D','B','C','D','B','D','E','F'))

df2 <- data.frame(variable_name = c(rep('var6', 4), rep('var4', 4)),
                  var_value = c(4,3,6,2,1,3,2,4),
                  value_type = c('A','B','C','D','B','D','E','F'))

ggplot(df1, aes(x = variable_name, y = var_value, fill = value_type)) + 
    geom_col(width = 0.8 * length(unique(df1$variable_name)) / 6)

ggplot(df2, aes(x = variable_name, y = var_value, fill = value_type)) + 
    geom_col(width = 0.8 * length(unique(df2$variable_name)) / 6)