Assignment 8

library(tidyverse)
library(sf)
library(tigris)
options(tigris_class = "sf")
library(tmap)

#Question 1

Calculate the Education, Housing, Health/Environment, Civic Life and overall ROI indices.

download.file(url = "https://raw.githubusercontent.com/crd150/data/master/assign8files.zip", destfile = "assign8files.zip")
unzip(zipfile = "assign8files.zip")

capitaltracts <- st_read("capitaltracts.shp", stringsAsFactors = FALSE)

## Reading layer `capitaltracts' from data source `/Users/sherigudez/Documents/CRD 150/Lab 8/capitaltracts.shp' using driver `ESRI Shapefile'
## Simple feature collection with 486 features and 39 fields
## geometry type:  POLYGON
## dimension:      XY
## bbox:           xmin: -122.422 ymin: 38.01842 xmax: -119.8772 ymax: 39.3165
## epsg (SRID):    NA
## proj4string:    +proj=longlat +ellps=GRS80 +no_defs

capitaltracts <- mutate(capitaltracts, Education= (edplc1+edplc2+edplc3+edplc4)/4, Housing= (hsplc1+hsplc2)/2, HealthEnv= (enplc1+enplc2+enplc3+enplc4)/4, Civic= (soplc1+soplc2)/2, Overall= (Education+Housing+HealthEnv+Civic)/4)

Using the quintile-based breaks to categorize tracts as Lowest Opportunity, Low, Moderate, High, and Highest Opportunity that we used in the lab guide, map the Education, Housing, Health/Environment, Civic Life, and overall ROI indices for the Capital region. You should have five separate maps for the four domains and the overall index

capitaltracts <- mutate(capitaltracts, EducationQ= cut(Education, breaks= quantile(Education, c(0,0.2,0.4,0.6,0.8,1), na.rm= TRUE), labels= c("Lowest Opportunity", "Low", "Moderate", "High", "Highest Opportunity")), HealthEnvQ= cut(HealthEnv, breaks= quantile(HealthEnv, c(0,0.2,0.4,0.6,0.8,1), na.rm= TRUE), labels= c("Lowest Opportunity", "Low", "Moderate", "High", "Highest Opportunity")), HousingQ= cut (Housing, breaks= quantile(Housing, c(0,0.2,0.4,0.6,0.8,1), na.rm= TRUE), labels= c("Lowest Opportunity", "Low", "Moderate", "High", "Highest Opportunity")), CivicQ= cut(Civic, breaks= quantile(Civic, c(0,0.2,0.4,0.6,0.8,1), na.rm= TRUE), labels= c("Lowest Opportunity", "Low", "Moderate", "High", "Highest Opportunity")), OverallQ= cut(Overall, breaks= quantile(Overall, c(0,0.2,0.4,0.6,0.8,1), na.rm= TRUE), labels= c("Lowest Opportunity", "Low", "Moderate", "High", "Highest Opportunity")))

tm_shape(capitaltracts)+
  tm_polygons(col= "OverallQ", palette= "Reds", border.alpha = 0, title= "Overall Opportunity", midpoint= NA)

tm_shape(capitaltracts)+
  tm_polygons(col= "EducationQ", palette= "Greens", border.alpha = 0, title= "Education Opportunity", midpoint= NA)

tm_shape(capitaltracts)+
  tm_polygons(col= "HousingQ", palette= "Purples", border.alpha = 0, title= "Housing Opportunity", midpoint= NA)

tm_shape(capitaltracts)+
  tm_polygons(col= "CivicQ", palette= "Blues", border.alpha = 0, title= "Civic Opportunity", midpoint= NA)

tm_shape(capitaltracts)+
  tm_polygons(col= "HealthEnvQ", palette= "Greys", border.alpha = 0, title= "Health Opportunity", midpoint= NA)

Map percent of residents under 18 year olds for the capital region using style = quantile. Based on a visual comparisons of the maps, do you think that youth are located in high overall opportunity places?

tm_shape(capitaltracts)+
  tm_polygons(col= "p18und", palette= "Purples", border.alpha = 0, title= "Children Population", midpoint= NA)

#Based on the maps I created, the highest population of children are found in areas with low to moderate Educational opportunity- with the highest concentration of children in an area with low educational opportunity. Likewise, the majority of the children ppulation can be found in areas with low to moderate Housing opportunity. For Civic Opportunity, areas with the most population of children can be found in areas with various levels of civic opportunity, but the higher concentrations canbe found in areas with the lowest civic opportunities. For Health Opportunity, half of the majority of the Capital's children reside in areas with high Health Opportuniy. But, the other half reside in areas with low to moderate Health Opportunity. The highest concentration of the Capital's children reside in an area with low Health Opportunity. Overall, children tend to reside in areas with low to moderate Overall Opportunity, with a few concentrations of them living in areas with the highest Overall Opportunity.

Calculate the correlation between percent under 18 year olds and all of the indices (overall and the 5 domains). Summarize your results.

summarize(capitaltracts,
          educccorr= cor(Education, p18und, use= "complete.obs"),
          housecorr= cor(Housing, p18und, use= "complete.obs"),
          healthenvcorr= cor(HealthEnv, p18und, use= "complete.obs"),
          civiccorr= cor(Civic, p18und, use= "complete.obs"),
          roicorr= cor(Overall, p18und, use= "complete.obs"))

## Simple feature collection with 1 feature and 5 fields
## geometry type:  POLYGON
## dimension:      XY
## bbox:           xmin: -122.422 ymin: 38.01842 xmax: -119.8772 ymax: 39.3165
## epsg (SRID):    NA
## proj4string:    +proj=longlat +ellps=GRS80 +no_defs
##    educccorr housecorr healthenvcorr   civiccorr    roicorr
## 1 -0.1111788 0.2521986    -0.3557642 -0.02018818 -0.1002093
##                         geometry
## 1 POLYGON ((-121.8625 38.0679...

#The results from my correlational tests show that there is a low to nonexistent negative correlation for Educational Opportunity (-0.11_ and percent under 18 years old, a low to nonexistent positive correlation between Housing Opportunity (0.2) and percent under 18 years. This same low to nonexistent correlational values can be seen in Health Opportunity, Civic Opportunity, and Overall Opportunity.

Calculate the correlation between percent black, percent Asian, and percent Hispanic and the overall ROI index. Summarize your results.

summarize(capitaltracts,
          roiblk= cor(Overall, pblk, use= "complete.obs"),
          roiasn= cor(Overall, pasn, use= "complete.obs"),
          roishsp= cor(Overall, phisp, use= "complete.obs"))

## Simple feature collection with 1 feature and 3 fields
## geometry type:  POLYGON
## dimension:      XY
## bbox:           xmin: -122.422 ymin: 38.01842 xmax: -119.8772 ymax: 39.3165
## epsg (SRID):    NA
## proj4string:    +proj=longlat +ellps=GRS80 +no_defs
##       roiblk      roiasn    roishsp                       geometry
## 1 -0.4492105 -0.07883642 -0.4762791 POLYGON ((-121.8625 38.0679...

#Based on my correlational analysis on racial demographicss and Overall Opportunity in California's Capital Region, you can see that the association between Percent Black and Overall Opportunity is a low to low moderate negative correlation (-.45). For Percent Asian and Overall Opportunity, there is an extremely low or noneexistent negative correlation (-.08). For Percent Hispanic and Overall Opportunity (-.48), there is a low to low moderate negative correlation.

#Question 2

Scan the Task Force’s Opportunity Mapping Methodology and answer the following questions regarding how they constructed the index.

How many domains did the Task Force include? What are they? The Task Force included 3 domains: Economics, Environmental Quality, and Education
What were the three criteria they used to select the variables that were included to construct the domains? The three criterias to help construt the domains were: evidence of peer-reviewed research that the indicator is linked to improved life outcomes for low-income families, particularly children; Reliable data; and Publicly available data
What kinds of tracts did they exclude from the index? Task Force excluded census tracts that had unreliable data, prisoner populations that make up at least 75 percent of the population, and low-population density tracts that have lower than 15 people (or five households) per square mile.
Did they take into account regional differences when calculating the index? If so, how? Task Force took account for regional differences by identifying higher resource areas within each region.
Explain why they incorporated a Poverty and Racial Segregation filter. Task Force incorporated a Poverty and Racial Segregation filter tas a tool to use to avoid further segregation and to increase access to to opportunity for low-income families
For 2020, the state is planning to make some changes to the map methodology. Read their summary of changes. In your own words, succinctly summarize these changes and their reasons. These mapping changes are being made to help inform new policy to incentive the government to build more housing opportunity areas in places with higher resources because, historically, the State has built low-income housing in segregated areas with low-resources.

The data contain the index values for all the domains (Economic: economicc, Environment: environc, Education educc) and the composite index (composite with the five categories category and labels label) for urban tracts in the Capital Region. Merge this file with the file used in Question 1 (capitaltracts) using the GEOID as the ID. Map the State’s overall opportunity index (category) using the appropriate labels.

state <- read_csv("https://raw.githubusercontent.com/crd150/data/master/lihtccapital.csv")

capitaltracts <- left_join(capitaltracts, state, by= "GEOID")

 tm_shape(capitaltracts) +
  tm_polygons("category",  palette = "Greens", breaks = c(0,1,2,3,4, 5), labels = c("High Segregation & Poverty", "Low Resource", "Moderate Resource", "High Resource", "Highest Resource"),
              border.alpha = 0, title = "State Opportunity", midpoint = NA)

Assignment 8

CRD 150

SHERIDAMAE GUDEZ

2/28/20