CODEWITHSUNDEEP
rggplot2bar-chart
Mrmoleje
Mrmoleje

Reputation: 493

ggplot2 bar chart labels and colours

I'm having an issue with adding labels and colours to a bar chart using geom_text.

Here's an eg of the data: Data

Season   Answer   n     freq         
Spring   Yes      103    0.77                    
Spring   No       30     0.23   
Winter   Yes      75     0.85
Winter   No       13     0.15

For labels

The labels bunch up together, rather than there being a figure at the end of each bar.

(chart example)

ggplot(data = a, aes(x = Answer, y = freq)) + 
    geom_bar(aes(fill = season),stat = "identity", position = "dodge") +
    theme_minimal() + 
    scale_y_continuous(labels = scales::percent, limits = c(0, 1)) +
    geom_text(aes(label = freq, group = Answer),
              position=position_dodge(width = 0.5), vjust = -1.5) +
    ggtitle(label = "x") +
    labs (x = "%") +
    coord_flip()

I would like there to be a proportion at the end of each bar, rather than them overlapping on each other.

I would also like the proportions to show as *100. So 77.0%, rather than 0.77

For colours

I would like to amend the colours from the standard blue and red here as well. When I add a palette with four colours each bar gets an individual colour (chart example 2), rather than one for 'spring' and one for 'winter'. You'll see that doing this also messes up all the labels and the legend.

If I use a colour palette with two colours I get this: 

Error: Aesthetics must be either length 1 or the same as the data (4): fill, x, y

ggplot(data = a, aes(x = Answer, y = freq)) + 
    geom_bar(aes(fill = "palette"),stat = "identity", position = "dodge") +
    theme_minimal() + 
    scale_y_continuous(labels=scales::percent,limits= c(0, 1))+
    geom_text(aes(label = freq, group = Answer),
              position = position_dodge(width = 0.5),
              vjust = -1.5) +
    ggtitle(label = "x") +
    labs (x = "%") +
    coord_flip()

Upvotes: 1

Views: 3730

Answers (1)

alistaire
alistaire

Reputation: 43334

To fix the text dodging, drop the group aesthetic and adjust the dodge amount. To set the fill palette, add a scale_fill_* call, e.g.

library(ggplot2)

a <- data.frame(Season = c("Spring", "Spring", "Winter", "Winter"), 
                Answer = c("Yes", "No", "Yes", "No"), 
                n = c(103L, 30L, 75L, 13L), 
                freq = c(0.77, 0.23, 0.85, 0.15), stringsAsFactors = FALSE)

ggplot(data = a, aes(x = Answer, y = freq, fill = Season, label = scales::percent(freq))) + 
    geom_col(position = "dodge") +
    geom_text(position = position_dodge(width = .9)) +
    scale_y_continuous(labels = scales::percent, limits = c(0, 1)) +
    scale_fill_brewer(type = 'qual') + 
    theme_minimal() + 
    labs(title = "x", x = "%") +
    coord_flip()

Upvotes: 1

Related Questions