Homicide Rates from Gender Perspective: Analysis using Radar Chart and Bootstrap Intervals

[This article was first published on DataGeeek, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

The violence in the regions is essential to indicate the peace and security reached by the countries. Fortunately, the global homicide rate has been decreasing while it is slowly. But as for men, the situation does not look so bright. The global homicide rate per 100.000 people is about four times higher for men when compared to women.

First, we will examine this situation using a radar chart for the year 2020.


#Intentional homicides, female (per 100,000 female)
df_vi_fe <- 
  WDI(indicator = "VC.IHR.PSRC.FE.P5", 
      extra = TRUE) %>% 
  as_tibble() %>% 
  mutate(gender = "female") %>% 
  rename(rate = VC.IHR.PSRC.FE.P5)

#Intentional homicides, male (per 100,000 male)
df_vi_ma <- 
  WDI(indicator = "VC.IHR.PSRC.MA.P5", 
      extra = TRUE) %>% 
  as_tibble() %>% 
  mutate(gender = "male") %>% 
  rename(rate = VC.IHR.PSRC.MA.P5)

#Combining all the datasets
df_merged <- 
  df_vi_fe %>% 
  rbind(df_vi_ma) %>% 
  #removing labels attribute for fitting process
  crosstable::remove_labels() %>% 

#The data frame of the international homicide rate by gender, 2020
df_2020 <- 
  df_merged %>% 
  filter(year == 2020,
         region != "Aggregates") %>% 
  select(region, gender, income, rate)

#Radar/spider chart

#Preparing the radar data frame for fmsb package
df_radar <- 
  df_2020 %>% 
  group_by(region, gender) %>% 
  summarise(mean = mean(rate)) %>% 
  pivot_wider(names_from = "region", 
              values_from = "mean") %>% 

#Adding the max and min of each variable to use the fmsb package
df_radar <- rbind(rep(32,7),

#Plotting the average homicide rates(per 100.000 people) 
#by gender in the Regions, 2020
           pcol = c("orange","steelblue"))

#Setting font family
par(family = "Bricolage Grotesque") 

#Plot title
title("Average Homicide Rates by Gender in the Regions, 2020", 
      sub = "(per 100.000 people)",
      font = 2)

legend(x= 0.7, 
       y= 1.2,
       legend = c("Female", "Male"), 
       bty = "n", 
       pch=20 , 
       text.col = "black", 

As you can see from the above chart, Latin America & the Caribbean had the highest average number (per 100.000 people) of male homicides in 2020 by far; this could be related to organized crime, which is common in the area.

Now, we will model the homicide rates of the regions, with and without gender, using bootstrap confidence intervals to understand the motives behind it.

#Bootstrap intervals

without_gender <- 
  reg_intervals(rate ~ region + income, 
                data = df_2020, 
                times = 500)

with_gender <- 
  reg_intervals(rate ~ region + income + gender, 
                data = df_2020, 
                times = 500)

#Bootstrap confidence intervals plot

#Legend colors for the title
legend_cols <- RColorBrewer::brewer.pal(3, "Dark2")

  without_gender %>% mutate(gender = "without"), 
  with_gender %>% mutate(gender = "with")
) %>%
  mutate(term = str_remove_all(term, "gender|income|region")) %>%
  mutate(term = str_to_title(term)) %>% 
             term %>% reorder(.estimate), 
             color = gender)) +
  geom_vline(xintercept = 0, 
             linewidth = 1.5, 
             lty = 2, 
             color = "gray50") +
  geom_errorbar(size = 1.4, 
                alpha = 0.7,
                aes(xmin = .lower, 
                    xmax = .upper)) +
  geom_point(size = 3) +
  scale_x_continuous() +
  scale_color_brewer(palette = "Dark2") +
  labs(x = "Higher indicates more important",
       y = "",
       title = glue::glue("Bootstrap Intervals <span style='color:{legend_cols[1]}'>with</span> or <span style='color:{legend_cols[2]}'>without</span> Gender")) +
  theme_minimal(base_family = "Bricolage Grotesque",
                base_size = 15) +
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.background = element_rect(fill = "#eaf7fa"),
        axis.title.x = element_text(size = 12),
        plot.title = ggtext::element_markdown(hjust = 0.5, face = "bold"))

Passing the vertical dashed line (zero point) in the related intervals indicates significantly not the importance of the related variables, which confirms the spider chart above for the Male and Latin America & Caribbean variables.

To leave a comment for the author, please follow the link and comment on their blog: DataGeeek.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)