2020 California net migration visualization using a flow map

california-migration-flowmap.md

I once had a discussion with a colleague and he was mentioning about Californians migrating to different states in the US. This led me to think about how I could visualize California migration. Through Google search, I stumbled upon this artice from USAFacts that presents facts on 2020 California migration using data from the IRS based on filed tax returns. Additionally, I also came across an online tool called FlowmapBlue that allows users to create beautiful flow maps to visualize movements happening across pairs of origin and destination locations.

I used the data from USAFacts, wrangled it using R to create the Google Sheet that FlowmapBlue expects, and published it as a flow map that can be accessed here. As my colleague rightly pointed out, it seems Californians are avoiding Midwest due to the cold weather!

flowmap.R

# Net migration to California in 2020
# Source: https://usafacts.org/articles/725000-people-left-california-in-2020-which-states-did-they-move-to/
cali_net_migration_2020 <- tibble::tribble(
            ~State,    ~Net,
        "New York",   6031L,
        "Illinois",   1978L,
      "New Jersey",   1311L,
   "Massachusetts",    537L,
  "Washington, DC",    161L,
    "Rhode Island",   -191L,
    "North Dakota",   -210L,
       "Louisiana",   -229L,
          "Alaska",   -277L,
        "Delaware",   -310L,
   "West Virginia",   -365L,
     "Connecticut",   -442L,
         "Vermont",   -539L,
     "Mississippi",   -584L,
   "New Hampshire",   -865L,
        "Maryland",   -910L,
       "Minnesota",   -984L,
        "Nebraska",  -1033L,
           "Maine",  -1122L,
       "Wisconsin",  -1189L,
            "Iowa",  -1255L,
        "Michigan",  -1344L,
    "Pennsylvania",  -1607L,
         "Wyoming",  -1669L,
    "South Dakota",  -1669L,
          "Kansas",  -1712L,
          "Hawaii",  -2201L,
            "Ohio",  -2217L,
        "Kentucky",  -2362L,
         "Indiana",  -2444L,
         "Alabama",  -2502L,
      "New Mexico",  -2505L,
        "Arkansas",  -4428L,
        "Virginia",  -4524L,
         "Montana",  -4813L,
        "Missouri",  -4920L,
  "South Carolina",  -5034L,
        "Oklahoma",  -6137L,
         "Georgia",  -8872L,
  "North Carolina", -11681L,
            "Utah", -11964L,
        "Colorado", -12618L,
          "Oregon", -17109L,
       "Tennessee", -18201L,
      "Washington", -18762L,
         "Florida", -20867L,
           "Idaho", -21558L,
          "Nevada", -30386L,
         "Arizona", -37825L,
           "Texas", -69342L
  )

# Create flows data
flows <- cali_net_migration_2020 |>
  dplyr::mutate(
    origin = dplyr::if_else(Net > 0, State, "California"),
    dest = dplyr::if_else(Net > 0, "California", State),
    count = abs(Net)
  ) |>
  dplyr::select(origin, dest, count) |>
  dplyr::mutate(
    origin = state.abb[match(origin, state.name)],
    origin = dplyr::coalesce(origin, "DC"),
    dest = state.abb[match(dest, state.name)]
  )
dplyr::glimpse(flows)

# Create locations data
locations <- cali_net_migration_2020 |>
  dplyr::select(State) |>
  dplyr::filter(State != "Washington, DC") |>
  dplyr::add_row(State = "California") |>
  tidygeocoder::geocode(
    state = State, method = "osm", lat = "lat", long = "lon"
  ) |>
  dplyr::bind_rows(
    tibble::tibble(
      State = "Washington, DC"
    ) |>
      tidygeocoder::geocode(
        city = State, method = "osm", lat = "lat", long = "lon"
      )
  ) |>
  dplyr::rename(name = State) |>
  dplyr::mutate(
    id = state.abb[match(name, state.name)],
    id = dplyr::coalesce(id, "DC"),
    .before = 1
  )
dplyr::glimpse(locations)

# Prepare properties sheet
properties <- tibble::tribble(
                       ~property, ~value,
                         "title", "2020 California net migration",
                   "description", "Which states did Californians move to in 2020?",
                   "source.name", "USAFacts",
                    "source.url", "https://usafacts.org/articles/725000-people-left-california-in-2020-which-states-did-they-move-to/",
                "createdBy.name", "Ashirwad Barnwal",
               "createdBy.email", "[email protected]",
                 "createdBy.url", "https://www.linkedin.com/in/ashirwad1992/",
            "mapbox.accessToken", NA,
               "mapbox.mapStyle", NA,
                 "colors.scheme", "Default",
               "colors.darkMode", "yes",
                 "animate.flows", "yes",
                    "clustering", "yes",
                  "flows.sheets", "flows",
  "msg.locationTooltip.incoming", "Picked California",
  "msg.locationTooltip.outgoing", "Left California",
  "msg.locationTooltip.internal", "Internal & round migration",
    "msg.flowTooltip.numOfTrips", "Total migration",
       "msg.totalCount.allTrips", "{0} migration",
   "msg.totalCount.countOfTrips", "{0} of {1} migration"
)

# Create a new Google spreadsheet with properties, locations, and flows sheets
googlesheets4::gs4_create(
  name = "california-migration-flowmap",
  sheets = list(properties = properties, locations = locations, flows = flows)
)

flows.csv

locations.csv