Positioning the legend on ggplot

[Lilip]

In this example graph, the count of each group (or 'type') appears three times below the percentage of each result. What I am trying to do is to show the count ('total') just once per type, and to place it to the left, below the y-label, underneath A, B, and C. Alternatively, the count could be shown to the right of the bar. Do you have any suggestions on how I can achieve this?

df <- data.frame(
  type = c("A", "A","A", "B", "B","B", "C", "C","C"),
  result = c("negative", "positive", "negative", "unknow", "positive", "positive", 
            "negative","positive", "unknow" ))

myplot_a <- df %>%
  count(type, result) %>%       
  group_by(type) %>%
  mutate(pct= prop.table(n) * 100,
         total = sum(n)) %>% # add a column for total count
  
  ggplot() + aes(y= type, pct, fill=result) +
  geom_bar(stat="identity") +
  ylab("Type") +
  labs(fill = "", x = "Percentage") +
  scale_fill_manual(values=c("negative"="#00b3b3", "unknow"="#4d4d4d", "positive"="#e60000")) +
  geom_text(aes(label=paste0(sprintf("%1.1f", pct),"%\n", total)),
            position=position_stack(vjust=0.5)) +
  
theme_bw()+
  theme(text = element_text(size = 15)) #font size

myplot_a

[nirgrahamuk]

df <- data.frame(
  type = c("A", "A","A", "B", "B","B", "C", "C","C"),
  result = c("negative", "positive", "negative", "unknow", "positive", "positive", 
             "negative","positive", "unknow" ))

(my_prepped_data <-  df %>%
  count(type, result) %>% 
  arrange(type,desc(result)) %>%      
  group_by(type) %>%
  mutate(pct= prop.table(n),
         total = sum(n) ,
  pct_text = scales::label_percent()(pct),
         ttext=ifelse(row_number()==1,
                      paste0(pct_text,"\n",total),
                      paste0(pct_text,"\n",""))))

myplot_a <- my_prepped_data%>% 
  ggplot() + aes(y= type, pct, fill=result) +
  geom_bar(stat="identity") +
  ylab("Type") +
  labs(fill = "", x = "Percentage") +
  scale_fill_manual(values=c("negative"="#00b3b3", "unknow"="#4d4d4d", "positive"="#e60000")) +
  geom_text(aes(label=ttext),
            position=position_stack(vjust=0.5)) +
  
  theme_bw()+
  theme(text = element_text(size = 15)) #font size

myplot_a

[Lilip]

Thanks!
Would it be possible to display this value outside the bar, either to the left, underneath the 'A', 'B', and 'C' labels, or to the right, in front of the bars?

[nirgrahamuk]

df <- data.frame(
  type = c("A", "A","A", "B", "B","B", "C", "C","C"),
  result = c("negative", "positive", "negative", "unknow", "positive", "positive", 
             "negative","positive", "unknow" ))

(my_prepped_data <-  df %>%
    count(type, result) %>% 
    arrange(type,desc(result)) %>%      
    group_by(type) %>%
    mutate(pct= prop.table(n),
           total = sum(n) ,
           pct_text = scales::label_percent()(pct),
           type=paste0(type,"\n(",total,")")
))

myplot_a <- my_prepped_data%>% 
  ggplot() + aes(y= type, pct, fill=result) +
  geom_bar(stat="identity") +
  ylab("Type") +
  labs(fill = "", x = "Percentage") +
  scale_fill_manual(values=c("negative"="#00b3b3", "unknow"="#4d4d4d", "positive"="#e60000")) +
  geom_text(aes(label=pct_text),
            position=position_stack(vjust=0.5)) +
  theme_bw()+
  theme(text = element_text(size = 15)) #font size

myplot_a

[Lilip]

It worked. Thank you!