GeoSpatialProj

Redlining in Winston-Salem, NC

Introduction

First I will load all my packages I will need for this project.

library(leaflet)
library(tidyverse)

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ ggplot2   3.4.4     ✔ tibble    3.2.1
✔ lubridate 1.9.3     ✔ tidyr     1.3.1
✔ purrr     1.0.2     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(sf)

Linking to GEOS 3.11.0, GDAL 3.5.3, PROJ 9.1.0; sf_use_s2() is TRUE

library(readr)
library(mapview)
library(leaflet.providers)

According to law.cornell.edu (https://www.law.cornell.edu/wex/redlining), redlining is “Redlining can be defined as a discriminatory practice that consists of the systematic denial of services such as mortgages, insurance loans, and other financial services to residents of certain areas, based on their race or ethnicity.” In this project, I have decided to use redlining data in Winston-Salem, NC and compare it to the most recent census data of Winston-Salem. I would like to evaluate if there has been any changes to the or growth in the distribution of income and its correlation race. I hypothesize there should be some changes in areas of towns that have been industrialized throughout the years. I also expect that there will be a better distributions of ethnicity and races. I have started by downloading and importing the redlining data from the database at the University of Richmond.

redline <- st_read("~/Desktop/WSRedlining/mappinginequality (1).json") 

Reading layer `mappinginequality (1)' from data source 
  `/Users/LandreyMessick/Desktop/WSRedlining/mappinginequality (1).json' 
  using driver `GeoJSON'
Simple feature collection with 10154 features and 11 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: -122.7675 ymin: 25.70537 xmax: -69.60044 ymax: 48.2473
Geodetic CRS:  WGS 84

Next I filtered out the city of Winston-Salem from this data and made a map to better visualize the grades of each zone that the data provided in Winston-Salem.

redline |> 
  filter(state == "NC" & city == "Winston-Salem") ->wsredlines

ws_color <- colorFactor(c('green','blue', 'yellow', 'red'), wsredlines$grade)

redline |> 
  filter(state == "NC" & city == "Winston-Salem") ->wsredlines

redline_map <- leaflet() |> 
  addProviderTiles(providers$Esri.NatGeoWorldMap) |> 
  setView(-80.258922,36.093640,  zoom=12) |> 
  addPolygons(
    data = wsredlines, 
    color = ~ws_color(wsredlines$grade),
    fillColor = wsredlines$fill,
    fillOpacity = .8,
    weight = 1,
    stroke = FALSE,
    label = paste0(wsredlines$grade)) |> 
    addLegend("bottomright", pal = ws_color, values = wsredlines$grade,
      title = "Redlining Grade",
      opacity = .8
      )
redline_map

In this data, I found that the lower graded zones are focused more towards the center of the city. Higher graded zones/tracts are located in suburbs of Winston-Salem.

To compare this data to a more current depiction of income in Winston-Salem, I am downloading and importing the censusapi data to find the income of each tract in Winston-Salem.

devtools::install_github("hrecht/censusapi")

Skipping install of 'censusapi' from a github remote, the SHA1 (74334d4d) has not changed since last install.
  Use `force = TRUE` to force installation

library(censusapi)
Sys.setenv(CENSUS_KEY='a88f6d2102d71503ca55a2a329ec4efe543a8dc5')
Sys.getenv("CENSUS_KEY")

[1] "a88f6d2102d71503ca55a2a329ec4efe543a8dc5"

I am now filtering out the data to be specifically about Forsyth County, which is the county Winston-Salem is located in. I also filtered out for the data to be showing the average income of each tract it will show. I then made a map to better visualize this.

poverty_rate <- getCensus(
  name = "timeseries/poverty/saipe",
  vars = c("NAME", "SAEPOVRTALL_PT"),
  region = "county:067",
  regionin = "state:37",
  time = "from 2010")

library(tidycensus)

forsyth <- get_acs(
  state = "NC",
  county = "Forsyth",
  geography = "tract",
  variables = "B19013_001",
  geometry = TRUE,
  year = 2020
)

forsyth |> 
  sf::st_transform('+proj=longlat +datum=WGS84')  -> forsyth
pal <- colorBin("Oranges", domain = forsyth$estimate, 6, pretty=TRUE)


ws_color2 <- colorFactor(c('green', 'blue','red','yellow'), forsyth$estimate)
  
  modern_map <- leaflet() %>%
    addTiles() %>%
    setView(-80.258922,36.093640,  zoom=10) |> 
    addPolygons(data = forsyth,
                color = ~ws_color2(forsyth$estimate),
                fillColor = ~ws_color2(forsyth$estimate),
                fillOpacity = .8,
                weight = 1,
                label = paste0("$", forsyth$estimate),
                labelOptions = labelOptions(style = list("font-weight" = "normal", padding = "3px 8px"),
                                            textsize = "15px",
                                            direction = "auto") )
  modern_map

After evaluating this data, I have noticed that there is not much of a change from the previous redlining data. This is the opposite of what i hypothesized. The places that were graded low in the redlining data still have a lower income in comparison to other tracts. The places with a higher rating on the redlining data still have a higher income in comparison to the other tracts.

Since redlining is highly correlated with being a discriminatory practice based on race and/or ethnicity, I wanted to evaluate the tracts in Winston-Salem with the lowest and highest incomes. I did this by using the censusapi code and filtering out for races and ethnicity in these specific tracts. I then made a bar chart these two tracts to have an easier understanding and visualization.

racevars <- c(White = "P2_005N", 
              Black = "P2_006N", 
              Asian = "P2_008N", 
              Hispanic = "P2_002N")
Forsyth2 <- get_decennial(
  geography = "tract",
  variables = racevars,
  state = "NC",
  county = "Forsyth County",
  geometry = TRUE,
  summary_var = "P2_001N",
  year = 2020,
  sumfile = "pl"
)                 

Getting data from the 2020 decennial Census

Downloading feature geometry from the Census website.  To cache shapefiles for use in future sessions, set `options(tigris_use_cache = TRUE)`.

Using the PL 94-171 Redistricting Data Summary File

Note: 2020 decennial Census data use differential privacy, a technique that
introduces errors into data to preserve respondent confidentiality.
ℹ Small counts should be interpreted with caution.
ℹ See https://www.census.gov/library/fact-sheets/2021/protecting-the-confidentiality-of-the-2020-census-redistricting-data.html for additional guidance.
This message is displayed once per session.

Forsyth2 |> 
  filter(NAME == "Census Tract 5, Forsyth County, North Carolina") -> Census_Tract_5

Census_Tract_5 |>
  ggplot(aes(reorder(variable, value),value, fill=value)) + geom_col() + labs (title = "Tract with lowest average income") +
  xlab('Demographic') +
  ylab('N')

Forsyth2 |> 
  filter(NAME == "Census Tract 25.02, Forsyth County, North Carolina") -> Census_Tract_25.02

Census_Tract_25.02 |>
  ggplot(aes(reorder(variable, value),value, fill=value)) + geom_col() + labs (title = "Tract with Highest Average Income") +
  xlab('Demographic') +
  ylab('N')

After completing these bar charts, it is evident that there has not been much change or growth with this subject in Winston-Salem. The lowest income tract is predominantly Black individuals and the highest income tract is predominantly white individuals. This shows that not much has changes in Winston-Salem since the 1930s in regards to race, ethnicity, and income.