Here is a sample dataset:

#sample data
df <- tibble(year=c(1,1,1,1,2,2,2,2,3,3,3,3),
             col=c("a", "b", "c", "d", "a", "b", "c", "d", "a", "b", "c", "d"), 
             freq=c(2,3,5,1,4,3,8,3,5,7,3,9))

I want to create a bar plot for each year of data independently. I'd also like to print the total sample size for each year as a caption on the respective plot. I'm currently doing it manually like this:

#find total sample size for year 1
df_yr_1 <- df %>%
  filter(year==1) 
sum(df_yr_1$freq)

#make year 1 histogram
hist <- ggplot(df_yr_1, aes(x=col, y=freq))  
  geom_col()  
  labs(caption = "N=11")

hist

So on and so forth for each year.

Is there a way to automate this process? Ideally, all the plots would save to a pdf (1 per page), but if they're saved independently that's fine too. I have a feeling adding the total sample size as a caption might make the process more challenging, so if it's possible to automate the process without the sample size captions that would still be very helpful. Thanks in advance!

CodePudding user response：

A simple loop should do the trick:

pdf("my.pdf")
for(i in 1:3) {
  plot_df <- df %>% filter(year ==1)
  
  p <- ggplot(plot_df, aes(col, freq))  
    geom_col()  
    labs(title = paste("Year", i), caption = paste0("N=", sum(plot_df$freq)))
  
  print(p)
}
dev.off()

Resulting in

my.pdf

Page 1

Page 2

Page 3

CodePudding user response：

We could wrap the code in a loop

pdf( "plots.pdf", onefile = TRUE)
for(i in unique(df$year)) {
   df_yr_i <- df %>%
       filter(year==i) 
      


hist <- ggplot(df_yr_i, aes(x=col, y=freq))  
  geom_col()  
  labs(caption = paste0("N=", sum(df_yr_i$freq)))
  print(hist)
  

}

dev.off()

-output