Choropleth Map for Minnesota Counties Population in 2010

Data Wrangling

Step 1

Download the Shapefile data from Minnesota Geospatial Commons. The page was found via a Google search using the term Minnesota Shapefile. Use the read_sf function from the sf library to capture the data to a table. We will call it mn.

mn <- read_sf("./mn_shapefile_2/", "mn_county_boundaries_500")
head(mn)

## Simple feature collection with 6 features and 12 fields
## Geometry type: POLYGON
## Dimension:     XY
## Bounding box:  xmin: 190012.2 ymin: 5166465 xmax: 591752.2 ymax: 5472428
## Projected CRS: NAD83 / UTM zone 15N
## # A tibble: 6 × 13
##           AREA PERIMETER CTYONLY_ CTYONLY_ID  COUN CTY_NAME    CTY_ABBR CTY_FIPS
##          <dbl>     <dbl>    <dbl>      <dbl> <int> <chr>       <chr>       <int>
## 1  4608320924.   388250.        2          1    39 Lake of th… LOTW           77
## 2  2862183702.   263017.        3          2    35 Kittson     KITT           69
## 3  4347098503.   302591.        4          3    68 Roseau      ROSE          135
## 4  8167237871.   412897.        5          4    36 Koochiching KOOC           71
## 5  4698732288.   374208.        6          5    45 Marshall    MARS           89
## 6 17451037319.   682518.        7          6    69 St. Louis   STLO          137
## # ℹ 5 more variables: MaxSimpTol <dbl>, MinSimpTol <dbl>, Shape_Leng <dbl>,
## #   Shape_Area <dbl>, geometry <POLYGON [m]>

Step 2

Next, get the Minnesota population data in year 2010 to get more features like population, households, and persons per household.
From the Popfinder for Minnesota Counties, press Download all the data in this PopFinder tool. (Excel file)
Move the Excel file to the working folder (meaning the same folder as your R file) and name it to mn_county_bigdata.cvs.

Step 3

Clean the data by just considering the county geography type and the year 2010 from the data.

mn_clean <- read.csv("mn_county_bigdata.csv")
mn_clean <- mn_clean %>% 
  filter(Year==2010 & Geography.Type=="County")
head(mn_clean)

##   Geography.Type Geography.Name Year Population Households
## 1         County         Aitkin 2010    16,202      7,299 
## 2         County          Anoka 2010   330,844    121,227 
## 3         County         Becker 2010    32,504     13,224 
## 4         County       Beltrami 2010    44,442     16,846 
## 5         County         Benton 2010    38,451     15,079 
## 6         County      Big Stone 2010     5,269      2,293 
##   Persons.Per.Household..PPH.
## 1                        2.18
## 2                        2.70
## 3                        2.42
## 4                        2.51
## 5                        2.48
## 6                        2.24

Perform a left join to to our main data mn so it will have the additional features.

# Add Households and Persons Per Household variables to the main data table
mn <- mn %>% left_join(mn_clean, by = c("CTY_NAME" = "Geography.Name"))

Step 4

Now, let’s create categories for the population from very small to very large.

# Define the breakpoints for the categories for population
breakpoints <- c(0, 5000, 10000, 50000, 100000, Inf)


# Convert to integer
mn$Pop_int <- gsub(",", "", mn$Population) %>%
  as.integer()

# Create categories
mn$pop_category <- cut(mn$Pop_int, breaks = breakpoints, labels = c("very small (0-5,000)", "small (5,000 - 10,000)", "medium (10,000 - 50,000)", "large (50,000 - 100,000)", "very large (> 100,000)" ), include.lowest = TRUE)

This code chunk below combines the wanted features of a county into a character string. This helps as when you hover over a county on the map, it displays the correct information of name, population, households, and persons per household.

mn <- mn %>%
  mutate(info = paste(
    "Name: ", CTY_NAME,
    "\nPopulation Category: ", pop_category,
    "\nPopulation: ", Population,
    "\nHouseholds: ", Households,
    "\nPersons Per Household: ", Persons.Per.Household..PPH.
  ))

Step 5

Let’s plot a static choropleth map showing Minnesota counties by population in 2010 first. We use ggplot with passed in parameters below.

plot <- ggplot(mn, aes(fill = pop_category)) +
  geom_sf() + # call geom shapefile
  geom_sf_interactive(aes(geometry = geometry, tooltip = info)) + 
  scale_fill_manual(
    values = c(
      "very small (0-5,000)" = "#eff3ff",
      "small (5,000 - 10,000)" = "#bdd7e7",
      "medium (10,000 - 50,000)" = "#6baed6",
      "large (50,000 - 100,000)" = "#3182bd",
      "very large (> 100,000)" = "#08519c") ) +
  labs(title = "Minnesota counties by population category in 2010", 
       subtitle = "Most highly populated counties are on the East of MN",
       fill="",
       caption = "Long Truong (5/12/2023)\nSource: Minnesota Geospatial Commons and Demographic Center")  +
  theme(plot.caption = element_text(hjust = 0)) +
  theme(legend.position = "right") +
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank()) +
  theme(axis.text.x=element_blank(), 
      axis.ticks.x=element_blank(), 
      axis.text.y=element_blank(), 
      axis.ticks.y=element_blank()) 

plot

Choropleth map showing Minnesota counties by population in 2010

Choropleth Map for Minnesota Counties Population in 2010

Long Truong

Load packages