重新排列geom_bar中的栏

Question

I have a bar plot that I made using ggplot2 in R. I want to move some of the bars around. I've seen a few explanations for how to reorder by percentage, but I want to sort mine into a particular order that is determined by the name of the variables.

Here is my code:

# make OTU count data frame
count=c(Count_Litter, Count_02, Count_0210, Count_1020)
horizon=c('Litter', '2 cm', '2-10 cm', '10-20 cm')
count_data=data.frame(horizon, count)

# make bar chart
plot=ggplot(data=count_data, aes(x=horizon, y=count))
final_plot=plot + geom_bar(position='dodge', stat= 'identity', fill= 
    'red') + coord_flip() + geom_text(aes(label=count), hjust=1) + 
    labs(title='Number of OTUs by Horizon')

It gives:

I would like to switch the positions of the 2 cm bar and the 2 to 10 cm bar. So, from top to bottom, the y-axis should read: Litter, 2 cm, 2 to 10 cm, 10 to 20 cm. Any ideas?

Answer 1

count <- 1:4
#Change your horizon to a ordered factor and it should be fine.
horizon=factor(c('Litter', '2 cm', '2-10 cm', '10-20 cm'),
                   levels = c('Litter', '2 cm', '2-10 cm', '10-20 cm'),
                   ordered = T)

Answer 2

Creating data frame with a column for the specific order

# Library
library(ggplot2)

# Re-creating the data frame
count=c(556527, 132732, 129880, 148088)
horizon=c('Litter', '2 cm', '2-10 cm', '10-20 cm')
# Specifying the order in which the bar should appear
order=c(4, 3, 2, 1)
count_data=data.frame(horizon, count, order)

Specifying the specific order using the new column

# Specifying the levels
count_data$horizon <- factor(count_data$horizon, levels =             
count_data$horizon[order(count_data$order)])

Plotting

# making horizontal bar chart
plot=ggplot(data=count_data, aes(x=horizon, y=count))
plot + 
  geom_bar(position='dodge', stat= 'identity', fill= 'red') + 
  coord_flip() + 
  geom_text(aes(label=count), hjust=1) + 
  labs(title='Number of OTUs by Horizon')

Output https://raw.githubusercontent.com/magoavi/stackoverflow/master/50537609.png