1

I have this dataframe C_Em_df

structure(list(Driver = c("Crop agriculture", "Infrastructure", 
"Mining", "Mixed Agriculture", "Other land use", "Pasture", "Tree crops", 
"Water", "Crop agriculture", "Infrastructure", "Mining", "Mixed Agriculture", 
"Other land use", "Pasture", "Tree crops", "Water", "Crop agriculture", 
"Infrastructure", "Mining", "Mixed Agriculture", "Other land use", 
"Pasture", "Tree crops", "Water", "Crop agriculture", "Infrastructure", 
"Mining", "Mixed Agriculture", "Other land use", "Pasture", "Tree crops", 
"Water"), Period = c("1990-2000", "1990-2000", "1990-2000", "1990-2000", 
"1990-2000", "1990-2000", "1990-2000", "1990-2000", "1990-2000", 
"1990-2000", "1990-2000", "1990-2000", "1990-2000", "1990-2000", 
"1990-2000", "1990-2000", "2000-2005", "2000-2005", "2000-2005", 
"2000-2005", "2000-2005", "2000-2005", "2000-2005", "2000-2005", 
"2000-2005", "2000-2005", "2000-2005", "2000-2005", "2000-2005", 
"2000-2005", "2000-2005", "2000-2005"), Value = c(129536.358373574, 
14089.3660954917, 985.646531415156, 34951.5705930615, 75478.7796771996, 
1001024.77681633, 9673.51414314377, 51631.4446491193, 9.83294102032751, 
1.06950594852475, 0.0748191807457263, 2.65312948831128, 5.7294986378404, 
75.9865238911138, 0.73430421561273, 3.91927761752383, 273356.204972389, 
12040.5899468613, 607.505228212054, 45165.8223684273, 75748.9976185639, 
1221137.74328547, 8851.85933777376, 39629.016246337, 16.3048047540391, 
0.718181861746466, 0.0362357025480948, 2.69399377763239, 4.51818028644936, 
72.8368777437064, 0.527984496372407, 2.36374137750571), n = c("n = 1669", 
"n = 298", "n = 20", "n = 1355", "n = 1623", "n = 10986", "n = 316", 
"n = 466", "n = 1669", "n = 298", "n = 20", "n = 1355", "n = 1623", 
"n = 10986", "n = 316", "n = 466", "n = 783", "n = 151", "n = 7", 
"n = 925", "n = 851", "n = 6039", "n = 211", "n = 244", "n = 783", 
"n = 151", "n = 7", "n = 925", "n = 851", "n = 6039", "n = 211", 
"n = 244"), Type = c("Sum", "Sum", "Sum", "Sum", "Sum", "Sum", 
"Sum", "Sum", "Percentage", "Percentage", "Percentage", "Percentage", 
"Percentage", "Percentage", "Percentage", "Percentage", "Sum", 
"Sum", "Sum", "Sum", "Sum", "Sum", "Sum", "Sum", "Percentage", 
"Percentage", "Percentage", "Percentage", "Percentage", "Percentage", 
"Percentage", "Percentage")), .Names = c("Driver", "Period", 
"Value", "n", "Type"), row.names = c("1", "3", "5", "7", "9", 
"11", "13", "15", "12", "31", "51", "71", "91", "111", "131", 
"151", "2", "4", "6", "8", "10", "122", "14", "16", "21", "41", 
"61", "81", "101", "121", "141", "161"), class = "data.frame")

I want to use the parameter facet_grid in ggplot to plot in the same window one plot with the absolute values (Sum) and one plot with the percentage values (Percentage). To do that I need to have two different y-axis scales and also two different y-axis titles in the two plots. I have managed to do the line of code below, but could not get really what I want.

g <- ggplot(C_Em_df, aes(x = Driver, y = Value, fill = Period, width = .85)) +
  facet_grid(Type~., scales="free")+
geom_bar(position = "dodge", stat = "identity") +
  labs(x = "", y = "Carbon emission (T/Year)") +
  theme(axis.text = element_text(size = 16),
        axis.title = element_text(size = 20), 
        legend.title = element_text(size = 20, face = 'bold'),
        legend.text=  element_text(size=20),
        axis.line = element_line(colour = "black"))+
  scale_fill_grey("Period") +
  theme_classic(base_size = 20, base_family = "") + 
  theme(panel.grid.minor = element_line(colour="grey", size=0.5)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

enter image description here

hrbrmstr
  • 77,368
  • 11
  • 139
  • 205
Simon Besnard
  • 377
  • 1
  • 6
  • 18

1 Answers1

2

Here is the answer

#Plot absolute values
p1 <- ggplot(C_Em_df[C_Em_df$Type=="Sum",], aes(x = Driver, y = Value, fill = Period, width = .85)) +
  geom_bar(position = "dodge", stat = "identity") +
  labs(x = "", y = "Carbon emission (T/Year)") +
  theme(axis.text = element_text(size = 16),
        axis.title = element_text(size = 20), 
        legend.title = element_text(size = 20, face = 'bold'),
        legend.text=  element_text(size=20),
        axis.line = element_line(colour = "black"))+
  scale_fill_grey("Period") +
  theme_classic(base_size = 20, base_family = "") + 
  theme(panel.grid.minor = element_line(colour="grey", size=0.5)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

#add the number of observations
foo <- ggplot_build(p1)$data[[1]]
p2<-p1 + annotate("text", x = foo$x, y = foo$y + 50000, label = C_Em_df[C_Em_df$Type=="Sum",]$n, size = 4.5) 

#Plot Percentage values
p3 <- ggplot(C_Em_df[C_Em_df$Type=="Percentage",], aes(x = Driver, y = Value, fill = Period, width = .85)) +
  geom_bar(position = "dodge", stat = "identity") +
  scale_y_continuous(labels = percent_format(), limits=c(0,1))+
  labs(x = "", y = "Carbon Emissions (%)") +
  theme(axis.text = element_text(size = 16),
        axis.title = element_text(size = 20), 
        legend.title = element_text(size = 20, face = 'bold'),
        legend.text=  element_text(size=20),
        axis.line = element_line(colour = "black"))+
  scale_fill_grey("Period") +
  theme_classic(base_size = 20, base_family = "") + 
  theme(panel.grid.minor = element_line(colour="grey", size=0.5)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Plot two graphs together
install.packages("gridExtra")
library(gridExtra)
gA <- ggplotGrob(p2)
gB <- ggplotGrob(p3)
maxWidth = grid::unit.pmax(gA$widths[2:5], gB$widths[2:5])
gA$widths[2:5] <- as.list(maxWidth)
gB$widths[2:5] <- as.list(maxWidth)

p4 <- arrangeGrob(
  gA, gB, nrow = 2, heights = c(0.80, 0.80))
Simon Besnard
  • 377
  • 1
  • 6
  • 18
  • I just have one issue with the number of observations. When I label them they do not correspond to my variable (Driver per period). Do you how I can fix it? – Simon Besnard Jan 23 '15 at 15:10
  • If you look into `C_Em_df[C_Em_df$Type=="Sum",]`, you see how rows are like. You probably want to change the order of rows according to the levels of `Driver`. When you create `foo`, the df is reflecting the order of rows in `C_Em_df[C_Em_df$Type=="Sum",]`. That is why the labelling is not right at the moment. Hope this helps you. – jazzurro Jan 23 '15 at 15:33
  • An alternate method while still using facets - http://stackoverflow.com/questions/27929549/different-y-axis-labels-facet-grid-and-sizes?answertab=active#tab-top. But using separate plots and grid.arrange, as you have done, is more appropriate imo – user20650 Jan 23 '15 at 22:16