Cómo crear una pirámide de población con ggplot2

Problema

Queremos crear una pirámide de población con ggplot2. Nuestro ejemplo es la población española por edad y sexo en 2020.

Ejemplo

# A tibble: 42 x 3
   Age      Gender   Total
           
 1 "0-4 "   male   1018039
 2 "0-4 "   female  963653
 3 "5-9"    male   1196380
 4 "5-9"    female 1129063
 5 "10-14"  male   1297635
 6 "10-14"  female 1225863
 7 "15-19 " male   1232566
 8 "15-19 " female 1156455
 9 "20-24 " male   1207902
10 "20-24 " female 1152765
# ... with 32 more rows

df <- structure(list(Age = c("0-4 ", "0-4 ", "5-9", "5-9", "10-14", 
"10-14", "15-19 ", "15-19 ", "20-24 ", "20-24 ", "25-29 ", "25-29 ", 
"30-34 ", "30-34 ", "35-39 ", "35-39 ", "40-44 ", "40-44 ", "45-49 ", 
"45-49 ", "50-54 ", "50-54 ", "55-59 ", "55-59 ", "60-64 ", "60-64 ", 
"65-69 ", "65-69 ", "70-74 ", "70-74 ", "75-79 ", "75-79 ", "80-84 ", 
"80-84 ", "85-89 ", "85-89 ", "90-94 ", "90-94 ", "95-99 ", "95-99 ", 
"100+", "100+"), Gender = c("male", "female", "male", "female", 
"male", "female", "male", "female", "male", "female", "male", 
"female", "male", "female", "male", "female", "male", "female", 
"male", "female", "male", "female", "male", "female", "male", 
"female", "male", "female", "male", "female", "male", "female", 
"male", "female", "male", "female", "male", "female", "male", 
"female", "male", "female"), Total = c(1018039L, 963653L, 1196380L, 
1129063L, 1297635L, 1225863L, 1232566L, 1156455L, 1207902L, 1152765L, 
1308197L, 1275776L, 1421558L, 1417845L, 1702135L, 1688865L, 2024303L, 
1971909L, 1968659L, 1926866L, 1828015L, 1840434L, 1652558L, 1712299L, 
1410111L, 1502563L, 1153768L, 1270544L, 1020478L, 1191698L, 773823L, 
974046L, 513692L, 759379L, 361702L, 634714L, 133032L, 302885L, 
27305L, 84007L, 3732L, 13576L)), class = "data.frame", row.names = c(NA, 
-42L))
tibble(df)

Solución

Primero necesitamos crear dos columnas. En la primera cambiamos el signo de la población masculina para mostrarla invertida en el gráfico. En la segunda calculamos el porcentaje de población por edad y sexo.

df <- df %>%
  group_by(Gender) %>% 
  mutate(
    Population = ifelse(Gender == "female", Total,-Total),
    Percent = ifelse(Gender == "female", 100 * (Total / sum(Total)),-100 * (Total / sum(Total))))
df$Age <- factor(df$Age, levels=unique(df$Age)) # Para mantener el orden original de vector carácter

Totales

ggplot(df, aes(x = Age, Population, fill = Gender)) + 
  geom_bar(data = filter(df, Gender == "female"), stat = "identity") + 
  geom_bar(data = filter(df, Gender == "male"),  stat = "identity") + 
  scale_y_continuous(breaks = seq(-2000000, 2000000, 500000), labels = comma(abs(seq(-2000000, 2000000, 500000))))+
  coord_flip()

Porcentajes

ggplot(df, aes(x = Age, Percent, fill = Gender)) +
  geom_bar(data = filter(df, Gender == "female"), stat = "identity") +
  geom_bar(data = filter(df, Gender == "male"),  stat = "identity") +
  scale_y_continuous(breaks = seq(-10, 10, 2), labels = comma(abs(seq(-10, 10, 2)))) +
  coord_flip()

Referencias

• stackoverflow
• Principales series de población (INE)