Final Project

1. Introduction

Picture

Hate crimes are the highest priority of the FBI’s Civil Rights program due to the devastating impact they have on families and communities.

According to the article “Higher Rates Of Hate Crimes Are Tied To Income Inequality” by Maimuna Majumder, States with more income inequality are more likely to have higher rates of hate incidents before and after the Trump election.

In my project I would like to find the relationship between income inequality and hate crimes as well as to prove or deny the statement from the article. I also want to find out if Trump election influnced on hate crimes within the states.

2. Data

## 
## Attaching package: 'ggplot2'

## The following objects are masked from 'package:psych':
## 
##     %+%, alpha

## Warning: package 'dplyr' was built under R version 3.5.3

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Data collection:

Pre-election data collected by the FBI Uniform Crime Reporting Program from law enforcement agencies.

Post-election data collected by Southern Poverty Law Center that used media accounts and people’s self-reports.

Cases

## [1] 51 12

The data contains 51 cases (50 US States and District of Columbia) and 12 variables.

Variables

Share of the population who voted for Trump in 2016 is numerical variable

Gini index and Annual income are numerical variables.

Type of study

This is an observational study since there are no control and experimental groups.

3. Data Visualisation

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

##                   State Income
## 21             Maryland  76165
## 30        New Hampshire  73397
## 12               Hawaii  71223
## 7           Connecticut  70161
## 9  District of Columbia  68277

##            State Income
## 18      Kentucky  42786
## 19     Louisiana  42406
## 1        Alabama  42278
## 49 West Virginia  39552
## 25   Mississippi  35521

Now, let’s analyse how the income is distributed within the states and find the states with the highest income inequality.

The Gini Index is a summary measure of income inequality.The Gini coefficient ranges from 0, indicating perfect equality (where everyone receives an equal share), to 1, perfect inequality (where only one recipient or group of recipients receives all the income).

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

States with the Highest and Lowest Gini Index

##                   state    GI
## 9  District of Columbia 0.532
## 33             New York 0.499
## 7           Connecticut 0.486
## 19            Louisiana 0.475
## 22        Massachusetts 0.475

##            state    GI
## 47      Virginia 0.459
## 48    Washington 0.441
## 49 West Virginia 0.451
## 50     Wisconsin 0.430
## 51       Wyoming 0.423

Since the District of Columbia is not a state, but federal district, we will consider New York as the state with the highest income inequality (GI 0.499), followed by Connecticut (GI 0.486) and Louisiana (GI 0.475).

## [1] -0.1788214

The correlation between Median Income and GI is negative, meaning that in general states with the higher median income have lower GI.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

## Warning: Removed 1 rows containing non-finite values (stat_bin).

## Warning: Removed 1 rows containing non-finite values (stat_smooth).

## Warning: Removed 1 rows containing missing values (geom_point).

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

## Warning: Removed 4 rows containing non-finite values (stat_bin).

## Warning: Removed 4 rows containing non-finite values (stat_smooth).

## Warning: Removed 4 rows containing missing values (geom_point).

## 
## Call:
## lm(formula = hate_crimes_per_100k_splc ~ gini_index, data = dataset)
## 
## Coefficients:
## (Intercept)   gini_index  
##      -1.527        4.021

For every increase in the Gini Index of 0.1, the number of hate crimes per 100k people is predicted to increase by 0.4020510.

Let’s take a look at the relationship between the level of Trump support and Hate Crimes

## Warning: Removed 4 rows containing non-finite values (stat_boxplot).

4.Statistical Analysis

## Warning: Removed 4 rows containing non-finite values (stat_smooth).

## Warning: Removed 4 rows containing missing values (geom_point).

sharev1 <- lm(hate_crimes_per_100k_splc ~  share_voters_voted_trump, data = dataset)
summary(sharev1)

## 
## Call:
## lm(formula = hate_crimes_per_100k_splc ~ share_voters_voted_trump, 
##     data = dataset)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.33201 -0.11410 -0.02299  0.07858  0.56395 
## 
## Coefficients:
##                          Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                1.0173     0.1252   8.127 2.25e-10 ***
## share_voters_voted_trump  -1.4748     0.2522  -5.847 5.26e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.1926 on 45 degrees of freedom
##   (4 observations deleted due to missingness)
## Multiple R-squared:  0.4317, Adjusted R-squared:  0.4191 
## F-statistic: 34.19 on 1 and 45 DF,  p-value: 5.263e-07

The equation of the regression line:

\(Crimes=1.0173-1.4748*GI\)

When the share voters goes up by 0.1, we would expect the rate crimes goes down by 1.4748.

Looking at the p value for the t score, we can say that the share voters for Trump is statistically significant p(>|t|) = 5.263e-07.

Normality of the Residuals

##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -0.33201 -0.11410 -0.02299  0.00000  0.07858  0.56395

The distribution centers around 0, and the 1st and 2nd quantiles are close enough. The maximum and minimum values are not too far different.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

5. Conclusion

The initial question was to determine if there any relationship exists between the hate crimes and income inequality before and after the Trump election. Using plots and linear modeling we did confirm the statement from the article: “Higher Rates Of Hate Crimes Are Tied To Income Inequality” Also, Further data analysis proved that there are higher rate of hate crimes in the states with the lower share voters voted for Trump.

Resources