Final Project: Hate Crimes
Brian K. Liles
December 13, 2017
Hate-Crimes Data
#Import the hate crimes csv dataset
hate<-read.csv("C:\\Users\\lizza\\Documents\\CUNY - Data Analytics\\DATA 606 - Probablity and Statistics\\Final Project\\hate_crimes.csv", header=T)Dimensions & Names of the Hate dataset
Number of observations & variables
dim(hate)## [1] 51 12Variable Description & Names
Median household income,2016/ Share of the population that is unemployed (seasonally adjusted)/ Share of the population that lives in metropolitan areas,2015/ Share of adults 25 and older with a high-school degree, 2009/ Share of the population that are not U.S. citizens, 2015/ Share of white residents who are living in poverty, 2015/ Gini Index, 2015/ Share of the population that is not white, 2015/ Share of 2016 U.S. presidential voters who voted for Donal Trump/ Hate crimes per 100,000 population, Southern Poverty Law Center, Nov 9-18, 2016/ Average annual hate crimes per 100,000 population, FBI, 2010-2015/
names(hate)## [1] "state"
## [2] "median_household_income"
## [3] "share_unemployed_seasonal"
## [4] "share_population_in_metro_areas"
## [5] "share_population_with_high_school_degree"
## [6] "share_non_citizen"
## [7] "share_white_poverty"
## [8] "gini_index"
## [9] "share_non_white"
## [10] "share_voters_voted_trump"
## [11] "hate_crimes_per_100k_splc"
## [12] "avg_hatecrimes_per_100k_fbi"View the Heads & Tails of the Dataset
head(hate)## state median_household_income share_unemployed_seasonal
## 1 Alabama 42278 0.060
## 2 Alaska 67629 0.064
## 3 Arizona 49254 0.063
## 4 Arkansas 44922 0.052
## 5 California 60487 0.059
## 6 Colorado 60940 0.040
## share_population_in_metro_areas share_population_with_high_school_degree
## 1 0.64 0.821
## 2 0.63 0.914
## 3 0.90 0.842
## 4 0.69 0.824
## 5 0.97 0.806
## 6 0.80 0.893
## share_non_citizen share_white_poverty gini_index share_non_white
## 1 0.02 0.12 0.472 0.35
## 2 0.04 0.06 0.422 0.42
## 3 0.10 0.09 0.455 0.49
## 4 0.04 0.12 0.458 0.26
## 5 0.13 0.09 0.471 0.61
## 6 0.06 0.07 0.457 0.31
## share_voters_voted_trump hate_crimes_per_100k_splc
## 1 0.63 0.12583893
## 2 0.53 0.14374012
## 3 0.50 0.22531995
## 4 0.60 0.06906077
## 5 0.33 0.25580536
## 6 0.44 0.39052330
## avg_hatecrimes_per_100k_fbi
## 1 1.8064105
## 2 1.6567001
## 3 3.4139280
## 4 0.8692089
## 5 2.3979859
## 6 2.8046888tail(hate)## state median_household_income share_unemployed_seasonal
## 46 Vermont 60708 0.037
## 47 Virginia 66155 0.043
## 48 Washington 59068 0.052
## 49 West Virginia 39552 0.073
## 50 Wisconsin 58080 0.043
## 51 Wyoming 55690 0.040
## share_population_in_metro_areas
## 46 0.35
## 47 0.89
## 48 0.86
## 49 0.55
## 50 0.69
## 51 0.31
## share_population_with_high_school_degree share_non_citizen
## 46 0.910 0.01
## 47 0.866 0.06
## 48 0.897 0.08
## 49 0.828 0.01
## 50 0.898 0.03
## 51 0.918 0.02
## share_white_poverty gini_index share_non_white share_voters_voted_trump
## 46 0.10 0.444 0.06 0.33
## 47 0.07 0.459 0.38 0.45
## 48 0.09 0.441 0.31 0.38
## 49 0.14 0.451 0.07 0.69
## 50 0.09 0.430 0.22 0.48
## 51 0.09 0.423 0.15 0.70
## hate_crimes_per_100k_splc avg_hatecrimes_per_100k_fbi
## 46 0.3241491 1.9030814
## 47 0.3632489 1.7247546
## 48 0.6774876 3.8177403
## 49 0.3286771 2.0370536
## 50 0.2261971 1.1219447
## 51 NA 0.2669408Run basic descriptive statistics on the following variables: median_household_income, hate_crimes_per_100k_splc, avg_hatecrimes_per_100k_fbi
Median Household Income, 2016
summary(hate$median_household_income)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 35521 48657 54916 55224 60719 76165Hate Crimes Per 100,000 residents, 2016
summary(hate$hate_crimes_per_100k_splc)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.06745 0.14271 0.22620 0.30409 0.35694 1.52230 4Average Annual Hate Crimes per 100,000 residents, 2010-2105
summary(hate$avg_hatecrimes_per_100k_fbi)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.2669 1.2931 1.9871 2.3676 3.1843 10.9535 1Explore how many states there are which has a higher median income than the median household income
table(hate$median_household_income>54916)##
## FALSE TRUE
## 26 25Sort data to identify which states lead in several categories
5 States w/the Lowest Median Household Incomes
#create a subset of data
income <- subset(hate, select = c(state,median_household_income))
#sort by median household income
income_sort <- income[order(income$median_household_income),]
head(income_sort,n=5)## state median_household_income
## 25 Mississippi 35521
## 49 West Virginia 39552
## 1 Alabama 42278
## 19 Louisiana 42406
## 18 Kentucky 42786The 5 States w/the Highest Median Household Incomes
tail(income_sort,n=5)## state median_household_income
## 9 District of Columbia 68277
## 7 Connecticut 70161
## 12 Hawaii 71223
## 30 New Hampshire 73397
## 21 Maryland 76165We can conclude that Maryland has the largest median household income with 76165 while Mississippi has the smallest median household income with 35521.
The 5 States w/the Lowest Rate of White Poverty
#create a subset of data
poverty <- subset(hate, select = c(state,share_white_poverty))
#sort by states with populations of white poverty
poverty_sort <- poverty[order(poverty$share_white_poverty),]
head(poverty_sort,n=5)## state share_white_poverty
## 9 District of Columbia 0.04
## 24 Minnesota 0.05
## 2 Alaska 0.06
## 7 Connecticut 0.06
## 21 Maryland 0.06The 5 States w/the Highest Rate of White Poverty
tail(poverty_sort,n=5)## state share_white_poverty
## 20 Maine 0.12
## 43 Tennessee 0.13
## 25 Mississippi 0.14
## 49 West Virginia 0.14
## 18 Kentucky 0.17The District of Columbia had the lowest population of whites living in poverty with 0.04 while Kentucky had the highest population with 0.17.
The 5 States w/the Lowest Population of Trump Voters
#create a subset of data
trump <- subset(hate, select = c(state,share_voters_voted_trump))
#sort by states with populations of white poverty
trump_sort <- trump[order(trump$share_voters_voted_trump),]
head(trump_sort,n=5)## state share_voters_voted_trump
## 9 District of Columbia 0.04
## 12 Hawaii 0.30
## 5 California 0.33
## 46 Vermont 0.33
## 22 Massachusetts 0.34The 5 States w/the Highest Population of Trump Voters
tail(trump_sort,n=5)## state share_voters_voted_trump
## 18 Kentucky 0.63
## 35 North Dakota 0.64
## 37 Oklahoma 0.65
## 49 West Virginia 0.69
## 51 Wyoming 0.70The District of Columbia had the lowest population of Trump voters with 0.04 while Wyoming had the highest population with 0.70.
Create two datasets Where the Median Household Incomes are Above/Below 54916
Histograms for Lower/Higher States w/Non-Citizens
lower_income <- subset(hate, median_household_income < 54916)
higher_income <- subset(hate, median_household_income > 54916)Compare Non-Citizens in Lower Income States & Higher Income States
hist(lower_income$share_non_citizen,
#border=NA,
col="beige",#Or use: col=colors()[18]
main="Population of Non-Citizens in Lower Income States",
xlab="Population")
hist(higher_income$share_non_citizen,
#border=NA,
col="beige",#Or use: col=colors()[18]
main="Population of Non-Citizens in Higher Income States",
xlab="Population")
Creating Box Plots for Hate Crimes Per 100K
boxplot(lower_income$hate_crimes_per_100k_splc,
col="beige",
notch=T,
horizontal=T,
main="Hate Crimes Per 100K Residents",
xlab="Population of Lower Income States")
boxplot(higher_income$hate_crimes_per_100k_splc,
col="beige",
notch=T,
horizontal=T,
main="Hate Crimes Per 100K Residents",
xlab="Population of Higher Income States")
Creating Box Plots for Average Hate Crimes Per 100K
boxplot(lower_income$avg_hatecrimes_per_100k_fbi,
col="beige",
notch=T,
horizontal=T,
main="Average Hate Crimes Per 100K/FBI Statistics",
xlab="Population of Lower Income States")
boxplot(higher_income$avg_hatecrimes_per_100k_fbi,
col="beige",
notch=T,
horizontal=T,
main="Average Hate Crimes Per 100K/FBI Statistics",
xlab="Population of Higher Income States")
As per research, the Gini coefficien is a measure of statistical dispersion intended to represent the income or wealth distribution of a nations residents, and is the most commonly used measure of inequality.
Evaluating the Normal Distribution of the Gini Index
Gini Index
qqnorm(lower_income$gini_index, main="Gini Index: Lower Population")
qqline(lower_income$gini_index)
qqnorm(higher_income$gini_index, main="Gini Index: Higher Population")
qqline(higher_income$gini_index)
Trump Voters
qqnorm(lower_income$share_voters_voted_trump, main="Lower Income Population: Trump Voters")
qqline(lower_income$share_voters_voted_trump)
qqnorm(higher_income$share_voters_voted_trump, main="Higher Income Population: Trump Voters")
qqline(higher_income$share_voters_voted_trump)
Adults Over 25 w/High School Diploma
qqnorm(lower_income$share_population_with_high_school_degree, main="Lower Income Population: AdultsOver 25 w/HS Degree")
qqline(lower_income$share_population_with_high_school_degree)
#Confidence Intervals
Average Hate Crimes, FBI
fbi <- hate$avg_hatecrimes_per_100k_fbi
samp_fbi <- sample(fbi, 20)
sample_mean_fbi <-mean(samp_fbi)
se <-sd(samp_fbi)/sqrt(20)
lower <-sample_mean_fbi - 1.96 * se
upper <-sample_mean_fbi + 1.96 * se
c(lower,upper)## [1] 1.282219 3.307578Median Household Income
mhi <- hate$median_household_income
samp_mhi <- sample(mhi, 20)
sample_mean_mhi <-mean(samp_mhi)
se <-sd(samp_mhi)/sqrt(20)
lower <-sample_mean_mhi - 1.96 * se
upper <-sample_mean_mhi + 1.96 * se
c(lower,upper)## [1] 52110.25 59617.85Linear Regression
The correlation between the median household income and the population of adults with a high school degree
cor(hate$median_household_income,hate$share_population_with_high_school_degree)## [1] 0.65349inc_deg <-lm(hate$share_population_with_high_school_degree ~ hate$median_household_income, data=hate)
summary(inc_deg)##
## Call:
## lm(formula = hate$share_population_with_high_school_degree ~
## hate$median_household_income, data = hate)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.075845 -0.017330 -0.000675 0.018116 0.048849
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.356e-01 2.240e-02 32.846 < 2e-16 ***
## hate$median_household_income 2.418e-06 4.001e-07 6.043 2.01e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.02605 on 49 degrees of freedom
## Multiple R-squared: 0.427, Adjusted R-squared: 0.4154
## F-statistic: 36.52 on 1 and 49 DF, p-value: 2.007e-07plot(hate$share_population_with_high_school_degree ~ hate$median_household_income)
abline(inc_deg)