Tidyverse
Heleine Fouda
2023-11-11
Getting started
This assignment leverages most of the capabilities built in the tidyverse package.
Importing the data
Url <-read_csv( "https://raw.githubusercontent.com/fivethirtyeight/data/master/hate-crimes/hate_crimes.csv")## Rows: 51 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): state
## dbl (11): median_household_income, share_unemployed_seasonal, share_populati...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.spec(Url) ## cols(
## state = col_character(),
## median_household_income = col_double(),
## share_unemployed_seasonal = col_double(),
## share_population_in_metro_areas = col_double(),
## share_population_with_high_school_degree = col_double(),
## share_non_citizen = col_double(),
## share_white_poverty = col_double(),
## gini_index = col_double(),
## share_non_white = col_double(),
## share_voters_voted_trump = col_double(),
## hate_crimes_per_100k_splc = col_double(),
## avg_hatecrimes_per_100k_fbi = col_double()
## )data <- UrlData preparation
Removing missing values
hate_crimes <- data[complete.cases(data),]
hate_crimes## # A tibble: 45 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alabama 42278 0.06 0.64
## 2 Alaska 67629 0.064 0.63
## 3 Arizona 49254 0.063 0.9
## 4 Arkansas 44922 0.052 0.69
## 5 Califor… 60487 0.059 0.97
## 6 Colorado 60940 0.04 0.8
## 7 Connect… 70161 0.052 0.94
## 8 Delaware 57522 0.049 0.9
## 9 Distric… 68277 0.067 1
## 10 Florida 46140 0.052 0.96
## # ℹ 35 more rows
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>Data exploration
# Let's explore the data set with a scatter
ggplot(data = hate_crimes) +
geom_point(mapping = aes(x = state , y = avg_hatecrimes_per_100k_fbi, color = state, alpha = 0.5))
Let’s first get a sense of the hate crimes distribution with a histogram
ggplot(data = hate_crimes) +
geom_histogram (mapping = aes(x = avg_hatecrimes_per_100k_fbi)) + scale_x_log10() ## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Now let’s get a sense of the hate crimes distribution using a boxplot
ggplot(data = hate_crimes) +
geom_boxplot (mapping = aes(x = avg_hatecrimes_per_100k_fbi)) +
facet_wrap(~state)
# Distribution of household income
ggplot(hate_crimes, aes(x = median_household_income)) +
geom_histogram()+ scale_x_log10() ## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Descriptive statistics
Leveraging the deplyr, tydr, and janitor packages
glimpse(hate_crimes)## Rows: 45
## Columns: 12
## $ state <chr> "Alabama", "Alaska", "Arizona…
## $ median_household_income <dbl> 42278, 67629, 49254, 44922, 6…
## $ share_unemployed_seasonal <dbl> 0.060, 0.064, 0.063, 0.052, 0…
## $ share_population_in_metro_areas <dbl> 0.64, 0.63, 0.90, 0.69, 0.97,…
## $ share_population_with_high_school_degree <dbl> 0.821, 0.914, 0.842, 0.824, 0…
## $ share_non_citizen <dbl> 0.02, 0.04, 0.10, 0.04, 0.13,…
## $ share_white_poverty <dbl> 0.12, 0.06, 0.09, 0.12, 0.09,…
## $ gini_index <dbl> 0.472, 0.422, 0.455, 0.458, 0…
## $ share_non_white <dbl> 0.35, 0.42, 0.49, 0.26, 0.61,…
## $ share_voters_voted_trump <dbl> 0.63, 0.53, 0.50, 0.60, 0.33,…
## $ hate_crimes_per_100k_splc <dbl> 0.12583893, 0.14374012, 0.225…
## $ avg_hatecrimes_per_100k_fbi <dbl> 1.8064105, 1.6567001, 3.41392…dim(hate_crimes)## [1] 45 12str(hate_crimes)## tibble [45 × 12] (S3: tbl_df/tbl/data.frame)
## $ state : chr [1:45] "Alabama" "Alaska" "Arizona" "Arkansas" ...
## $ median_household_income : num [1:45] 42278 67629 49254 44922 60487 ...
## $ share_unemployed_seasonal : num [1:45] 0.06 0.064 0.063 0.052 0.059 0.04 0.052 0.049 0.067 0.052 ...
## $ share_population_in_metro_areas : num [1:45] 0.64 0.63 0.9 0.69 0.97 0.8 0.94 0.9 1 0.96 ...
## $ share_population_with_high_school_degree: num [1:45] 0.821 0.914 0.842 0.824 0.806 0.893 0.886 0.874 0.871 0.853 ...
## $ share_non_citizen : num [1:45] 0.02 0.04 0.1 0.04 0.13 0.06 0.06 0.05 0.11 0.09 ...
## $ share_white_poverty : num [1:45] 0.12 0.06 0.09 0.12 0.09 0.07 0.06 0.08 0.04 0.11 ...
## $ gini_index : num [1:45] 0.472 0.422 0.455 0.458 0.471 0.457 0.486 0.44 0.532 0.474 ...
## $ share_non_white : num [1:45] 0.35 0.42 0.49 0.26 0.61 0.31 0.3 0.37 0.63 0.46 ...
## $ share_voters_voted_trump : num [1:45] 0.63 0.53 0.5 0.6 0.33 0.44 0.41 0.42 0.04 0.49 ...
## $ hate_crimes_per_100k_splc : num [1:45] 0.1258 0.1437 0.2253 0.0691 0.2558 ...
## $ avg_hatecrimes_per_100k_fbi : num [1:45] 1.806 1.657 3.414 0.869 2.398 ...names(hate_crimes)## [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"# We can also quickly add our tally values to our tibble using add_tally().
hate_crimes %>%
add_tally() %>%
glimpse()## Rows: 45
## Columns: 13
## $ state <chr> "Alabama", "Alaska", "Arizona…
## $ median_household_income <dbl> 42278, 67629, 49254, 44922, 6…
## $ share_unemployed_seasonal <dbl> 0.060, 0.064, 0.063, 0.052, 0…
## $ share_population_in_metro_areas <dbl> 0.64, 0.63, 0.90, 0.69, 0.97,…
## $ share_population_with_high_school_degree <dbl> 0.821, 0.914, 0.842, 0.824, 0…
## $ share_non_citizen <dbl> 0.02, 0.04, 0.10, 0.04, 0.13,…
## $ share_white_poverty <dbl> 0.12, 0.06, 0.09, 0.12, 0.09,…
## $ gini_index <dbl> 0.472, 0.422, 0.455, 0.458, 0…
## $ share_non_white <dbl> 0.35, 0.42, 0.49, 0.26, 0.61,…
## $ share_voters_voted_trump <dbl> 0.63, 0.53, 0.50, 0.60, 0.33,…
## $ hate_crimes_per_100k_splc <dbl> 0.12583893, 0.14374012, 0.225…
## $ avg_hatecrimes_per_100k_fbi <dbl> 1.8064105, 1.6567001, 3.41392…
## $ n <int> 45, 45, 45, 45, 45, 45, 45, 4…Getting a quick summary of the data frame using the skimr package.
skim(hate_crimes)| Name | hate_crimes |
| Number of rows | 45 |
| Number of columns | 12 |
| _______________________ | |
| Column type frequency: | |
| character | 1 |
| numeric | 11 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| state | 0 | 1 | 4 | 20 | 0 | 45 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| median_household_income | 0 | 1 | 55299.49 | 8979.49 | 39552.00 | 48060.00 | 54916.00 | 60708.00 | 76165.00 | ▆▆▇▃▂ |
| share_unemployed_seasonal | 0 | 1 | 0.05 | 0.01 | 0.03 | 0.04 | 0.05 | 0.06 | 0.07 | ▃▇▇▇▂ |
| share_population_in_metro_areas | 0 | 1 | 0.78 | 0.16 | 0.34 | 0.69 | 0.81 | 0.90 | 1.00 | ▁▂▃▅▇ |
| share_population_with_high_school_degree | 0 | 1 | 0.87 | 0.03 | 0.80 | 0.84 | 0.87 | 0.89 | 0.92 | ▃▆▅▇▇ |
| share_non_citizen | 0 | 1 | 0.06 | 0.03 | 0.01 | 0.03 | 0.05 | 0.08 | 0.13 | ▇▇▆▃▂ |
| share_white_poverty | 0 | 1 | 0.09 | 0.02 | 0.04 | 0.07 | 0.09 | 0.10 | 0.17 | ▂▇▅▂▁ |
| gini_index | 0 | 1 | 0.46 | 0.02 | 0.42 | 0.44 | 0.46 | 0.47 | 0.53 | ▅▇▅▁▁ |
| share_non_white | 0 | 1 | 0.32 | 0.15 | 0.06 | 0.21 | 0.30 | 0.42 | 0.63 | ▃▇▅▅▂ |
| share_voters_voted_trump | 0 | 1 | 0.48 | 0.11 | 0.04 | 0.41 | 0.49 | 0.57 | 0.69 | ▁▁▆▇▆ |
| hate_crimes_per_100k_splc | 0 | 1 | 0.30 | 0.25 | 0.07 | 0.14 | 0.23 | 0.35 | 1.52 | ▇▂▁▁▁ |
| avg_hatecrimes_per_100k_fbi | 0 | 1 | 2.37 | 1.72 | 0.41 | 1.32 | 1.94 | 3.14 | 10.95 | ▇▃▁▁▁ |
## # see summary for specified columns
skim(msleep, genus, vore, sleep_total)| Name | msleep |
| Number of rows | 83 |
| Number of columns | 11 |
| _______________________ | |
| Column type frequency: | |
| character | 2 |
| numeric | 1 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| genus | 0 | 1.00 | 3 | 13 | 0 | 77 | 0 |
| vore | 7 | 0.92 | 4 | 7 | 0 | 4 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| sleep_total | 0 | 1 | 10.43 | 4.45 | 1.9 | 7.85 | 10.1 | 13.75 | 19.9 | ▅▅▇▆▂ |
# summarizing across
sum1 <- hate_crimes %>%
summarize(across(state:avg_hatecrimes_per_100k_fbi, mean, na.rm = TRUE))## Warning: There were 2 warnings in `summarize()`.
## The first warning was:
## ℹ In argument: `across(state:avg_hatecrimes_per_100k_fbi, mean, na.rm = TRUE)`.
## Caused by warning:
## ! The `...` argument of `across()` is deprecated as of dplyr 1.1.0.
## Supply arguments directly to `.fns` through an anonymous function instead.
##
## # Previously
## across(a:b, mean, na.rm = TRUE)
##
## # Now
## across(a:b, \(x) mean(x, na.rm = TRUE))
## ℹ Run `dplyr::last_dplyr_warnings()` to see the 1 remaining warning.sum1## # A tibble: 1 × 12
## state median_household_income share_unemployed_seasonal share_population_in_…¹
## <dbl> <dbl> <dbl> <dbl>
## 1 NA 55299. 0.0508 0.782
## # ℹ abbreviated name: ¹share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>summary(hate_crimes$avg_hatecrimes_per_100k_fbi)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.412 1.325 1.937 2.374 3.136 10.953# Summarizing the average hate crimes using the janitor package and the function tabyl
summ_crimes <- hate_crimes %>%
tabyl(avg_hatecrimes_per_100k_fbi)
summ_crimes ## avg_hatecrimes_per_100k_fbi n percent
## 0.4120118 1 0.02222222
## 0.4309276 1 0.02222222
## 0.5613956 1 0.02222222
## 0.6980703 1 0.02222222
## 0.7527683 1 0.02222222
## 0.8692089 1 0.02222222
## 1.0440158 1 0.02222222
## 1.0816721 1 0.02222222
## 1.1219447 1 0.02222222
## 1.2626798 1 0.02222222
## 1.2825718 1 0.02222222
## 1.3248395 1 0.02222222
## 1.3411696 1 0.02222222
## 1.4699796 1 0.02222222
## 1.6567001 1 0.02222222
## 1.7247546 1 0.02222222
## 1.7573566 1 0.02222222
## 1.8064105 1 0.02222222
## 1.8864352 1 0.02222222
## 1.8913305 1 0.02222222
## 1.9030814 1 0.02222222
## 1.9089550 1 0.02222222
## 1.9370828 1 0.02222222
## 2.0370536 1 0.02222222
## 2.1059886 1 0.02222222
## 2.1139902 1 0.02222222
## 2.1439867 1 0.02222222
## 2.3840650 1 0.02222222
## 2.3979859 1 0.02222222
## 2.6862484 1 0.02222222
## 2.8046888 1 0.02222222
## 2.9549594 1 0.02222222
## 3.1021643 1 0.02222222
## 3.1360512 1 0.02222222
## 3.2004423 1 0.02222222
## 3.2404204 1 0.02222222
## 3.3948861 1 0.02222222
## 3.4139280 1 0.02222222
## 3.6124118 1 0.02222222
## 3.7727015 1 0.02222222
## 3.8177403 1 0.02222222
## 4.2078896 1 0.02222222
## 4.4132026 1 0.02222222
## 4.8018993 1 0.02222222
## 10.9534797 1 0.02222222# Note, that tabyl assumes categorical variables. summary(hate_crimes$share_non_white)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0600 0.2100 0.3000 0.3176 0.4200 0.6300Let’s do some filtering using dplyr
# Let's filter data set rows to only include households with a median income equal or less than 50000
low_income_states <-hate_crimes %>%
dplyr::filter (median_household_income <= 45000)
low_income_states## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alabama 42278 0.06 0.64
## 2 Arkansas 44922 0.052 0.69
## 3 Kentucky 42786 0.05 0.56
## 4 Louisiana 42406 0.06 0.81
## 5 South Ca… 44929 0.057 0.79
## 6 Tennessee 43716 0.057 0.82
## 7 West Vir… 39552 0.073 0.55
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>low_income_states %>%
arrange(desc(median_household_income))## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 South Ca… 44929 0.057 0.79
## 2 Arkansas 44922 0.052 0.69
## 3 Tennessee 43716 0.057 0.82
## 4 Kentucky 42786 0.05 0.56
## 5 Louisiana 42406 0.06 0.81
## 6 Alabama 42278 0.06 0.64
## 7 West Vir… 39552 0.073 0.55
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>low_income_states## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alabama 42278 0.06 0.64
## 2 Arkansas 44922 0.052 0.69
## 3 Kentucky 42786 0.05 0.56
## 4 Louisiana 42406 0.06 0.81
## 5 South Ca… 44929 0.057 0.79
## 6 Tennessee 43716 0.057 0.82
## 7 West Vir… 39552 0.073 0.55
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl># The 7 states with the lowest household income
(slice_tail(low_income_states , n=7))## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alabama 42278 0.06 0.64
## 2 Arkansas 44922 0.052 0.69
## 3 Kentucky 42786 0.05 0.56
## 4 Louisiana 42406 0.06 0.81
## 5 South Ca… 44929 0.057 0.79
## 6 Tennessee 43716 0.057 0.82
## 7 West Vir… 39552 0.073 0.55
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>Let’s filter both specific observations or rows and specific cases or columns that are of interest to us
# Filter data set columns to only include ...
low_income_df <- low_income_states %>% select(state,avg_hatecrimes_per_100k_fbi, median_household_income,share_population_with_high_school_degree,share_white_poverty,share_non_white,share_voters_voted_trump, share_non_citizen)
low_income_df## # A tibble: 7 × 8
## state avg_hatecrimes_per_1…¹ median_household_inc…² share_population_wit…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alabama 1.81 42278 0.821
## 2 Arkansas 0.869 44922 0.824
## 3 Kentucky 4.21 42786 0.817
## 4 Louisiana 1.34 42406 0.822
## 5 South Ca… 1.94 44929 0.836
## 6 Tennessee 3.14 43716 0.831
## 7 West Vir… 2.04 39552 0.828
## # ℹ abbreviated names: ¹avg_hatecrimes_per_100k_fbi, ²median_household_income,
## # ³share_population_with_high_school_degree
## # ℹ 4 more variables: share_white_poverty <dbl>, share_non_white <dbl>,
## # share_voters_voted_trump <dbl>, share_non_citizen <dbl> # Let's filter data set rows to only include households with a median income equal or superior to 65000
middle_income_states <-hate_crimes %>%
dplyr::filter(median_household_income >= 66000) # The 7 states with the highest household income
(slice_head(middle_income_states , n=7))## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alaska 67629 0.064 0.63
## 2 Connecti… 70161 0.052 0.94
## 3 District… 68277 0.067 1
## 4 Maryland 76165 0.051 0.97
## 5 Minnesota 67244 0.038 0.75
## 6 New Hamp… 73397 0.034 0.63
## 7 Virginia 66155 0.043 0.89
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl># alternatively
arrange(middle_income_states)## # A tibble: 7 × 12
## state median_household_inc…¹ share_unemployed_sea…² share_population_in_…³
## <chr> <dbl> <dbl> <dbl>
## 1 Alaska 67629 0.064 0.63
## 2 Connecti… 70161 0.052 0.94
## 3 District… 68277 0.067 1
## 4 Maryland 76165 0.051 0.97
## 5 Minnesota 67244 0.038 0.75
## 6 New Hamp… 73397 0.034 0.63
## 7 Virginia 66155 0.043 0.89
## # ℹ abbreviated names: ¹median_household_income, ²share_unemployed_seasonal,
## # ³share_population_in_metro_areas
## # ℹ 8 more variables: share_population_with_high_school_degree <dbl>,
## # share_non_citizen <dbl>, share_white_poverty <dbl>, gini_index <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # hate_crimes_per_100k_splc <dbl>, avg_hatecrimes_per_100k_fbi <dbl>Now let’s take the original data set and then group it by states before re-arranging it in descending order based on their average crime rates
# Grouping the data set by state and re-arranging them in descending order based on their average hate crime rates
states <- hate_crimes %>%
group_by(state) %>%
select(state, avg_hatecrimes_per_100k_fbi) %>%
summarize(N=n(), mean_hatecrimes = avg_hatecrimes_per_100k_fbi) %>%
arrange(desc(mean_hatecrimes))
states## # A tibble: 45 × 3
## state N mean_hatecrimes
## <chr> <int> <dbl>
## 1 District of Columbia 1 11.0
## 2 Massachusetts 1 4.80
## 3 New Jersey 1 4.41
## 4 Kentucky 1 4.21
## 5 Washington 1 3.82
## 6 Connecticut 1 3.77
## 7 Minnesota 1 3.61
## 8 Arizona 1 3.41
## 9 Oregon 1 3.39
## 10 Ohio 1 3.24
## # ℹ 35 more rowsLet’s select the 7 states with the lowest average hate crime rates.
lowest_hcrimes<- states %>%
slice_tail(n = 7)
lowest_hcrimes ## # A tibble: 7 × 3
## state N mean_hatecrimes
## <chr> <int> <dbl>
## 1 Illinois 1 1.04
## 2 Arkansas 1 0.869
## 3 Texas 1 0.753
## 4 Florida 1 0.698
## 5 Iowa 1 0.561
## 6 Pennsylvania 1 0.431
## 7 Georgia 1 0.412sum_lowest_hcrimes <- lowest_hcrimes %>%
summarize(across(state:mean_hatecrimes, mean, na.rm = TRUE))## Warning: There was 1 warning in `summarize()`.
## ℹ In argument: `across(state:mean_hatecrimes, mean, na.rm = TRUE)`.
## Caused by warning in `mean.default()`:
## ! argument is not numeric or logical: returning NAsum_lowest_hcrimes ## # A tibble: 1 × 3
## state N mean_hatecrimes
## <dbl> <dbl> <dbl>
## 1 NA 1 0.681Let’s select the 7 states with the highest average hate crime rates, using dplyr
highest_hcrimes<- states %>%
slice_head(n = 7)
highest_hcrimes ## # A tibble: 7 × 3
## state N mean_hatecrimes
## <chr> <int> <dbl>
## 1 District of Columbia 1 11.0
## 2 Massachusetts 1 4.80
## 3 New Jersey 1 4.41
## 4 Kentucky 1 4.21
## 5 Washington 1 3.82
## 6 Connecticut 1 3.77
## 7 Minnesota 1 3.61sum_highest_crimes <- highest_hcrimes %>%
summarize(across(state:mean_hatecrimes, mean, na.rm = TRUE))## Warning: There was 1 warning in `summarize()`.
## ℹ In argument: `across(state:mean_hatecrimes, mean, na.rm = TRUE)`.
## Caused by warning in `mean.default()`:
## ! argument is not numeric or logical: returning NAsum_highest_crimes ## # A tibble: 1 × 3
## state N mean_hatecrimes
## <dbl> <dbl> <dbl>
## 1 NA 1 5.08Some states with low household income score high in hate crimes
Low_income_states <-low_income_df%>%
dplyr::filter(avg_hatecrimes_per_100k_fbi >= 20000536)
view(Low_income_states)Low income states with the highest average rate crime:
Low_income1 <-low_income_df%>%
dplyr::filter(avg_hatecrimes_per_100k_fbi >= 20000536) %>%
arrange(desc(avg_hatecrimes_per_100k_fbi))
Low_income1## # A tibble: 0 × 8
## # ℹ 8 variables: state <chr>, avg_hatecrimes_per_100k_fbi <dbl>,
## # median_household_income <dbl>,
## # share_population_with_high_school_degree <dbl>, share_white_poverty <dbl>,
## # share_non_white <dbl>, share_voters_voted_trump <dbl>,
## # share_non_citizen <dbl>Let’s explore the relationship between average hate crimes and the other variables within the data set. Which factor best explain hate crimes rate?
Controlling for education
# Controlling for secondary education
High_school <- hate_crimes %>%
group_by(state) %>%
select(share_population_with_high_school_degree, avg_hatecrimes_per_100k_fbi) %>%
arrange(desc(share_population_with_high_school_degree))## Adding missing grouping variables: `state` High_school ## # A tibble: 45 × 3
## # Groups: state [45]
## state share_population_with_high_school_degree avg_hatecrimes_per_1…¹
## <chr> <dbl> <dbl>
## 1 Minnesota 0.915 3.61
## 2 Alaska 0.914 1.66
## 3 Iowa 0.914 0.561
## 4 New Hampshire 0.913 2.11
## 5 Vermont 0.91 1.90
## 6 Montana 0.908 2.95
## 7 Utah 0.904 2.38
## 8 Nebraska 0.898 2.69
## 9 Wisconsin 0.898 1.12
## 10 Kansas 0.897 2.14
## # ℹ 35 more rows
## # ℹ abbreviated name: ¹avg_hatecrimes_per_100k_fbiThere is a positive but very weak relationship between hate crime rates and secondary education
cor(hate_crimes$share_population_with_high_school_degree, hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.1405676# Controlling for secondary education
ggplot(hate_crimes, aes(x = avg_hatecrimes_per_100k_fbi, y = share_population_with_high_school_degree, color = median_household_income))+
geom_point()
Controlling for the share of non white population: very weak positive
relationship
# controlling for the share of non white population
non_white <- hate_crimes %>%
group_by(state) %>%
select(share_non_white, avg_hatecrimes_per_100k_fbi) %>%
arrange(desc(share_non_white))## Adding missing grouping variables: `state` non_white ## # A tibble: 45 × 3
## # Groups: state [45]
## state share_non_white avg_hatecrimes_per_100k_fbi
## <chr> <dbl> <dbl>
## 1 District of Columbia 0.63 11.0
## 2 New Mexico 0.62 1.89
## 3 California 0.61 2.40
## 4 Texas 0.56 0.753
## 5 Maryland 0.5 1.32
## 6 Nevada 0.5 2.11
## 7 Arizona 0.49 3.41
## 8 Georgia 0.48 0.412
## 9 Florida 0.46 0.698
## 10 New Jersey 0.44 4.41
## # ℹ 35 more rowscor(hate_crimes$share_non_white, hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.1345048Controlling for share of non citizen: Weak positive relationship here but there seems to be an outlier represented by the District of Columbia
# Controlling for share of non citizen
non_citizen <- hate_crimes %>%
group_by(state) %>%
select(share_voters_voted_trump, share_white_poverty, share_non_citizen , avg_hatecrimes_per_100k_fbi) %>%
arrange(desc(share_non_citizen ))## Adding missing grouping variables: `state` non_citizen## # A tibble: 45 × 5
## # Groups: state [45]
## state share_voters_voted_t…¹ share_white_poverty share_non_citizen
## <chr> <dbl> <dbl> <dbl>
## 1 California 0.33 0.09 0.13
## 2 District of Col… 0.04 0.04 0.11
## 3 New Jersey 0.42 0.07 0.11
## 4 Texas 0.53 0.08 0.11
## 5 Arizona 0.5 0.09 0.1
## 6 Nevada 0.46 0.08 0.1
## 7 New York 0.37 0.1 0.1
## 8 Florida 0.49 0.11 0.09
## 9 Massachusetts 0.34 0.08 0.09
## 10 Georgia 0.51 0.09 0.08
## # ℹ 35 more rows
## # ℹ abbreviated name: ¹share_voters_voted_trump
## # ℹ 1 more variable: avg_hatecrimes_per_100k_fbi <dbl>cor(hate_crimes$share_non_citizen , hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.3125537ggplot(hate_crimes, aes(x= avg_hatecrimes_per_100k_fbi, y = share_non_citizen, color= median_household_income )) +
geom_point(alpha = 0.5)
controlling for household income: a weak but positive relationship
# controlling for household income
cor(hate_crimes$median_household_income , hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.2906101# controlling for household income
ggplot(hate_crimes, aes(x = avg_hatecrimes_per_100k_fbi, y = median_household_income, color = median_household_income))+
geom_point(alpha= 0.5)
facet_wrap(~median_household_income)## <ggproto object: Class FacetWrap, Facet, gg>
## compute_layout: function
## draw_back: function
## draw_front: function
## draw_labels: function
## draw_panels: function
## finish_data: function
## init_scales: function
## map_data: function
## params: list
## setup_data: function
## setup_params: function
## shrink: TRUE
## train_scales: function
## vars: function
## super: <ggproto object: Class FacetWrap, Facet, gg>Controlling for population in metropolitan areas: A weak but positive relationship
cor(hate_crimes$share_population_in_metro_areas , hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.21617ggplot(hate_crimes, aes(x = avg_hatecrimes_per_100k_fbi, y = share_population_in_metro_areas,color = median_household_income))+
geom_point(alpha = 0.5)
Controlling for share_white_poverty: There is a negative relationship between hate crimes and the level of white poverty
cor(hate_crimes$share_white_poverty, hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] -0.2426443ggplot(hate_crimes, aes(x = avg_hatecrimes_per_100k_fbi,
y =share_white_poverty, color= median_household_income))+
geom_point(alpha = 0.5)
Controlling for seasonal unemployment: A very weak but positive realtionship
cor(hate_crimes$share_unemployed_seasonal, hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] 0.1721765ggplot(hate_crimes, aes(x = avg_hatecrimes_per_100k_fbi, y= share_unemployed_seasonal, color = median_household_income))+
geom_point(alpha = 0.5)
controlling for Trump votes: a strong but negative relationship
cor(hate_crimes$share_voters_voted_trump, hate_crimes$avg_hatecrimes_per_100k_fbi)## [1] -0.5580764# controlling for Trump votes
voted_trump <- hate_crimes %>%
group_by(state) %>%
select(share_voters_voted_trump, share_white_poverty, avg_hatecrimes_per_100k_fbi) %>%
arrange(desc(share_voters_voted_trump))## Adding missing grouping variables: `state` voted_trump ## # A tibble: 45 × 4
## # Groups: state [45]
## state share_voters_voted_t…¹ share_white_poverty avg_hatecrimes_per_1…²
## <chr> <dbl> <dbl> <dbl>
## 1 West Virgi… 0.69 0.14 2.04
## 2 Oklahoma 0.65 0.1 1.08
## 3 Alabama 0.63 0.12 1.81
## 4 Kentucky 0.63 0.17 4.21
## 5 Tennessee 0.61 0.13 3.14
## 6 Arkansas 0.6 0.12 0.869
## 7 Nebraska 0.6 0.07 2.69
## 8 Idaho 0.59 0.11 1.89
## 9 Louisiana 0.58 0.12 1.34
## 10 Indiana 0.57 0.12 1.76
## # ℹ 35 more rows
## # ℹ abbreviated names: ¹share_voters_voted_trump, ²avg_hatecrimes_per_100k_fbiggplot(hate_crimes, aes(x =avg_hatecrimes_per_100k_fbi, y = share_voters_voted_trump, color = median_household_income)) +
geom_point(alpha = 0.5)
Findings and Conclusion
None of the variables taken alone fully explains the average hate
crime rates noticed in the data frame. This could point to the
explanation that these crimes are the result of a combination of
factors. Geographically, the states with the lowest average hate crime
rates are typically not “border” states. Many of them of these states
are located in the U.S. hinterland and don’t have/share an international
border. These states are: Illinois
Arkansas
Texas
Florida Iowa
Pennsylvania
Georgia
Likewise, states that experience the highest hate crime are typically
not the ones with lowest household income range, nor are they “border”
states. Besides New Jersey, many of these state are also located in the
U.S hinterland. And, except for Kentucky, many of these states are
affluent states, with substantial size of the population holding high
school degrees. These states are in descending order: District of
Columbia
Massachusetts
New Jersey
Kentucky
Washington
Connecticut Minnesota Finally, voting for Trump was not found to
increase or decrease hate crimes across states.