Mapmaking Exercise

Explore non-spatial data

kerala_covid_cases_df |> glimpse()

Rows: 56
Columns: 4
$ Wave     <chr> "First", "First", "First", "First", "First", "First", "First"…
$ District <chr> "Alappuzha", "Ernakulam", "Idukki", "Kannur", "Kasaragod", "K…
$ Cases    <dbl> 82076, 130735, 29449, 59719, 32277, 92640, 84556, 128970, 123…
$ Deaths   <dbl> 406, 456, 48, 320, 111, 338, 217, 508, 449, 177, 131, 878, 49…

Create a Total Population column

# Lets create a total population column
kerala_district_sf <- kerala_district_sf |> 
  mutate(top_pop = sum(pop_2020, na.rm = T))

Visualise the spatial files

kerala_district_sf |> 
  ggplot() +
  geom_sf()

Create a choropleth map based on the population of the district

kerala_district_sf |> 
  ggplot() +
  geom_sf(aes(fill = pop_2020)) +
  scale_fill_distiller(palette = "Spectral")

Join both the spatial and non-spatial data

# Step 1: Check if the district names are given correctly
# Check if the district names in both the datasets are the same
(kerala_covid_cases_df |> distinct(District) |> pull(District)) %in% (kerala_district_sf |> distinct(distname) |> pull(distname))

 [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE

# Step 2: Reshape your data
kerala_covid_cases_df_reshaped <- kerala_covid_cases_df |> 
  pivot_wider(names_from = Wave, values_from = c(Cases, Deaths))

# Step 3: Identify the common variable in both the datasets (here it is `distname` and `District`)
kerala_district_joined_sf <- left_join(kerala_district_sf, kerala_covid_cases_df_reshaped, by = c("distname" = "District"))

ASSIGNMENTS

1. Create a new variable named estimating the standarized incidence rate.

(SIR can be estimated by SIR = cases / ((total_cases / total_pop) * pop) )

joined_df <- kerala_covid_cases_df |> 
  left_join(kerala_district_sf |> st_drop_geometry(),
            by = c("District" = "distname")) |> 
  group_by(Wave) |> 
  mutate(SIR = Cases / ((sum(Cases)/ top_pop) * pop_2020)) |> 
  ungroup()

kerala_district_sf <- kerala_district_sf |> left_join(joined_df |> select(District, SIR, Wave) |> rename(distname = District))

Joining with `by = join_by(distname)`

kerala_district_sf |> 
  ggplot() + 
  geom_sf(aes(fill = SIR))  +
  scale_fill_distiller(palette = "Spectral") +
  facet_wrap(~Wave)

2. Create a new variable named estimating the standarized motality rate.

(SMR can be estimated by SMR = deaths / ((total_deaths / total_pop) * pop) )

# Write the code here

3. Plot a choropleth map of Cases for each of the wave

# Write the code here

4. Plot a choropleth map of Deaths for each of the wave

# Write the code here

5. Plot a choropleth map of SIR for each of the wave

# Write the code here

6. Plot a choropleth map of SMR for each of the wave

# Write the code here