Data Management Assignment
Lorraine Gaudio
2023-04-10
Introduction
Suicide in the US is a major public health issue. In 2020, there were more than 45,000 recorded suicides. Significant research has gone into understanding the risk and protective factors associated with suicide. In this assignment, we will use publicly available data to examine suicide rates by US county.
Each county in the US has an identifying 5-digit code. For instance, Autauga, Alabama has the code 01001. This county usually appears first alphabetically. This is the “key” we will use to merge our datafiles together.
Our goal is to determine which variables best predict county-level suicide of the following: population density (the number of people / sq mile of land area), income, percent of people who identify as white, and the percentage of people who live alone.
Previous research has shown that rurality is associated with suicide, we might predict that income affects suicide (although different hypotheses might predict different directions), we know that suicidality is patterned by ethnic category. People who identify as white or those who identify as Native American have higher rates of suicide than other ethnic groups in the US. We hypothesize that living alone might contribute to suicide.
Load Packages
library(readxl)
library(tidyverse)
library(dplyr)
library(dslabs)Data Munging
We will use the following datasets:
CDC suicide data 2010 through 2020.txtUSCensusBureau Land Area.xlsUSCensusIncome5Y2020.csvUSCensusHousehold5Y2020.csvUSCensusDemographic5Y2020.csv
Step 1 Load files
# Load .txt file
CDC_suicide_data <- read.table("CDC suicide data 2010 through 2020.txt", header = TRUE, sep = "\t", stringsAsFactors = FALSE)
# Load .csv files
USCensusIncome <- read.csv("USCensusIncome5Y2020.csv", stringsAsFactors = FALSE)
USCensusHousehold <- read.csv("USCensusHousehold5Y2020.csv", stringsAsFactors = FALSE)
USCensusDemographic <- read.csv("USCensusDemographic5Y2020.csv", stringsAsFactors = FALSE)
# Load .xls file
USCensusBureau_LandArea <- read_excel("USCensusBureau Land Area.xls")Step 2: Data Munging
Eventually we will merge the dataframes together. Before we do so, we will identify or create the columns of interest. We will likely find it easier to rename some columns to something easier to work with as well.
First, we determine the suicide rates by US county by using
CDC_suicide_data. It has data on the number of suicides
in each county for the 10 year period 2010 – 2020. If the number of
suicides is less than 20, the data states that the crude rate is
“Unreliable”. If the number of suicides is less than 10, then the actual
number is “Suppressed”.
head(CDC_suicide_data)## County County.Code Deaths Population Crude.Rate
## 1 Autauga County, AL 1001 115 609875 18.9
## 2 Baldwin County, AL 1003 444 2250866 19.7
## 3 Barbour County, AL 1005 41 287620 14.3
## 4 Bibb County, AL 1007 38 248120 15.3
## 5 Blount County, AL 1009 122 635351 19.2
## 6 Bullock County, AL 1011 11 114915 UnreliableIf the Crude.Rate equals “Unreliable”, then we need to manually
calculate it. This equation is Deaths /
Population x 100,000.
# Recalculate Crude.Rate where it is "Unreliable"
CDC_suicide_data <- CDC_suicide_data %>%
mutate(Crude.Rate = ifelse(Crude.Rate == "Unreliable", as.numeric(Deaths) / as.numeric(Population) * 100000, as.numeric(Crude.Rate)))
head(CDC_suicide_data)## County County.Code Deaths Population Crude.Rate
## 1 Autauga County, AL 1001 115 609875 18.900000
## 2 Baldwin County, AL 1003 444 2250866 19.700000
## 3 Barbour County, AL 1005 41 287620 14.300000
## 4 Bibb County, AL 1007 38 248120 15.300000
## 5 Blount County, AL 1009 122 635351 19.200000
## 6 Bullock County, AL 1011 11 114915 9.572293Unfortunately, the “key” or county code varies a bit from dataset to
dataset. How can we manage these so that we will be able to merge the
files together? We merge by County.Code but most of the
dataframes use the column name Geography so we change the
name to Geography. This column will also need to be
character class.
CDC_suicide_data <- CDC_suicide_data %>%
mutate(County.Code = as.character(County.Code)) %>%
rename(Geography = County.Code)
head(CDC_suicide_data)## County Geography Deaths Population Crude.Rate
## 1 Autauga County, AL 1001 115 609875 18.900000
## 2 Baldwin County, AL 1003 444 2250866 19.700000
## 3 Barbour County, AL 1005 41 287620 14.300000
## 4 Bibb County, AL 1007 38 248120 15.300000
## 5 Blount County, AL 1009 122 635351 19.200000
## 6 Bullock County, AL 1011 11 114915 9.572293Geography needs to be a 5 digit number.
# Ensure Geography in CDC_suicide_data has a consistent 5-digit format
CDC_suicide_data <- CDC_suicide_data %>%
mutate(Geography = stringr::str_pad(Geography, width = 5, pad = "0"))
head(CDC_suicide_data)## County Geography Deaths Population Crude.Rate
## 1 Autauga County, AL 01001 115 609875 18.900000
## 2 Baldwin County, AL 01003 444 2250866 19.700000
## 3 Barbour County, AL 01005 41 287620 14.300000
## 4 Bibb County, AL 01007 38 248120 15.300000
## 5 Blount County, AL 01009 122 635351 19.200000
## 6 Bullock County, AL 01011 11 114915 9.572293Next, we need to identify the needed data about the following variables:
population density (the number of people / sq mile of land area)
income
percent of people who identify as white
percentage of people who live alone.
1. Population Density
USCensusBureau_LandArea has information on the land area
of each county is square miles. The variable we are interested in is:
LND_SQMI. This will be useful for finding population
density.
head(USCensusBureau_LandArea)## # A tibble: 6 × 3
## Areaname STCOU LND_SQMI
## <chr> <chr> <dbl>
## 1 UNITED STATES 00000 3787425.
## 2 ALABAMA 01000 52423.
## 3 Autauga, AL 01001 604.
## 4 Baldwin, AL 01003 2027.
## 5 Barbour, AL 01005 905.
## 6 Bibb, AL 01007 626.We select columns STCOU, LND_SQMI, and
County.Name.
# Select columns in USCensusBureau_LandArea
USCensusBureau_LandArea_select <- USCensusBureau_LandArea %>%
select(STCOU, LND_SQMI )
# Rename 'STCOU' to 'Geography'
USCensusBureau_LandArea_select <- rename(USCensusBureau_LandArea_select, Geography = STCOU)
head(USCensusBureau_LandArea_select)## # A tibble: 6 × 2
## Geography LND_SQMI
## <chr> <dbl>
## 1 00000 3787425.
## 2 01000 52423.
## 3 01001 604.
## 4 01003 2027.
## 5 01005 905.
## 6 01007 626.We join the USCensusBureau_LandArea_select dataframe to
the CDC_suicide_data dataframe using
County.Name.
# Join the dataframes on 'Geography'
suicide_data <- left_join(CDC_suicide_data, USCensusBureau_LandArea_select, by = "Geography")
head(suicide_data)## County Geography Deaths Population Crude.Rate LND_SQMI
## 1 Autauga County, AL 01001 115 609875 18.900000 604.49
## 2 Baldwin County, AL 01003 444 2250866 19.700000 2027.08
## 3 Barbour County, AL 01005 41 287620 14.300000 904.59
## 4 Bibb County, AL 01007 38 248120 15.300000 625.50
## 5 Blount County, AL 01009 122 635351 19.200000 650.65
## 6 Bullock County, AL 01011 11 114915 9.572293 626.11Then, we create a new column population.density in
CDC_suicide_data by dividing Population by
LND_SQMI. We need to make sure that
population.density is not Inf. We do this by
including an ifelse function when population.density is
created so that it is missing if land area is 0.
# Create the 'population.density' column
suicide_data$population.density <- ifelse(suicide_data$LND_SQMI == 0,
NA,as.numeric(suicide_data$Population) / suicide_data$LND_SQMI)## Warning in ifelse(suicide_data$LND_SQMI == 0, NA,
## as.numeric(suicide_data$Population)/suicide_data$LND_SQMI): NAs introduced by
## coercionhead(suicide_data)## County Geography Deaths Population Crude.Rate LND_SQMI
## 1 Autauga County, AL 01001 115 609875 18.900000 604.49
## 2 Baldwin County, AL 01003 444 2250866 19.700000 2027.08
## 3 Barbour County, AL 01005 41 287620 14.300000 904.59
## 4 Bibb County, AL 01007 38 248120 15.300000 625.50
## 5 Blount County, AL 01009 122 635351 19.200000 650.65
## 6 Bullock County, AL 01011 11 114915 9.572293 626.11
## population.density
## 1 1008.9083
## 2 1110.3982
## 3 317.9562
## 4 396.6747
## 5 976.4866
## 6 183.53802. Income
USCensusIncome as information on income from American
Community Survey run by the US Census bureau. This data is a 5-year
estimate of income. The column:
Estimate..Households..Median.income..dollars. is likely the
one we will want to know.
head(USCensusIncome)## Geography Geographic.Area.Name Estimate..Households..Total
## 1 0500000US01001 Autauga County, Alabama 21559
## 2 0500000US01003 Baldwin County, Alabama 84047
## 3 0500000US01005 Barbour County, Alabama 9322
## 4 0500000US01007 Bibb County, Alabama 7259
## 5 0500000US01009 Blount County, Alabama 21205
## 6 0500000US01011 Bullock County, Alabama 3429
## Margin.of.Error..Households..Total
## 1 366
## 2 1143
## 3 338
## 4 299
## 5 430
## 6 195
## Estimate..Households..Total..Less.than..10.000
## 1 6.2
## 2 5.2
## 3 14.6
## 4 11.0
## 5 10.1
## 6 18.2
## Margin.of.Error..Households..Total..Less.than..10.000
## 1 1.4
## 2 1.0
## 3 2.6
## 4 3.3
## 5 2.0
## 6 4.6
## Estimate..Households..Total...10.000.to..14.999
## 1 4.6
## 2 4.8
## 3 7.6
## 4 7.1
## 5 4.6
## 6 9.8
## Margin.of.Error..Households..Total...10.000.to..14.999
## 1 1.2
## 2 0.7
## 3 1.5
## 4 2.8
## 5 1.0
## 6 5.0
## Estimate..Households..Total...15.000.to..24.999
## 1 12.3
## 2 7.7
## 3 18.4
## 4 10.5
## 5 11.0
## 6 11.9
## Margin.of.Error..Households..Total...15.000.to..24.999
## 1 2.0
## 2 0.8
## 3 2.2
## 4 2.7
## 5 1.9
## 6 4.6
## Estimate..Households..Total...25.000.to..34.999
## 1 8.5
## 2 10.0
## 3 9.4
## 4 9.1
## 5 11.2
## 6 10.6
## Margin.of.Error..Households..Total...25.000.to..34.999
## 1 1.7
## 2 1.1
## 3 1.7
## 4 2.7
## 5 1.7
## 6 4.0
## Estimate..Households..Total...35.000.to..49.999
## 1 12.7
## 2 14.0
## 3 12.5
## 4 11.2
## 5 14.5
## 6 8.0
## Margin.of.Error..Households..Total...35.000.to..49.999
## 1 1.9
## 2 1.4
## 3 1.9
## 4 3.0
## 5 1.9
## 6 3.9
## Estimate..Households..Total...50.000.to..74.999
## 1 17.0
## 2 16.8
## 3 16.3
## 4 19.4
## 5 17.9
## 6 17.4
## Margin.of.Error..Households..Total...50.000.to..74.999
## 1 2.0
## 2 1.4
## 3 2.9
## 4 3.9
## 5 2.2
## 6 5.3
## Estimate..Households..Total...75.000.to..99.999
## 1 13.4
## 2 13.7
## 3 7.3
## 4 15.6
## 5 10.8
## 6 7.5
## Margin.of.Error..Households..Total...75.000.to..99.999
## 1 2.3
## 2 1.2
## 3 1.6
## 4 3.7
## 5 1.5
## 6 2.9
## Estimate..Households..Total...100.000.to..149.999
## 1 16.4
## 2 15.5
## 3 9.1
## 4 11.2
## 5 12.4
## 6 12.4
## Margin.of.Error..Households..Total...100.000.to..149.999
## 1 2.3
## 2 1.0
## 3 2.0
## 4 2.5
## 5 1.7
## 6 4.1
## Estimate..Households..Total...150.000.to..199.999
## 1 4.8
## 2 5.8
## 3 2.2
## 4 2.8
## 5 4.9
## 6 1.3
## Margin.of.Error..Households..Total...150.000.to..199.999
## 1 1.3
## 2 0.7
## 3 0.8
## 4 1.5
## 5 1.0
## 6 2.0
## Estimate..Households..Total...200.000.or.more
## 1 4.2
## 2 6.6
## 3 2.6
## 4 2.2
## 5 2.7
## 6 2.8
## Margin.of.Error..Households..Total...200.000.or.more
## 1 1.2
## 2 0.9
## 3 1.2
## 4 1.3
## 5 0.8
## 6 1.8
## Estimate..Households..Median.income..dollars.
## 1 57982
## 2 61756
## 3 34990
## 4 51721
## 5 48922
## 6 33866
## Margin.of.Error..Households..Median.income..dollars.
## 1 4839
## 2 2268
## 3 2909
## 4 6237
## 5 2269
## 6 10094
## Estimate..Households..Mean.income..dollars.
## 1 75614
## 2 83626
## 3 51557
## 4 61655
## 5 66360
## 6 50664
## Margin.of.Error..Households..Mean.income..dollars.
## 1 5718
## 2 2634
## 3 4229
## 4 5787
## 5 3864
## 6 5359Let’s remove “0500000US” off the Geography column.
# Remove "0500000US" from the 'Geography' column in USCensusIncome
USCensusIncome$Geography <- gsub("0500000US", "", USCensusIncome$Geography)
head(USCensusIncome)## Geography Geographic.Area.Name Estimate..Households..Total
## 1 01001 Autauga County, Alabama 21559
## 2 01003 Baldwin County, Alabama 84047
## 3 01005 Barbour County, Alabama 9322
## 4 01007 Bibb County, Alabama 7259
## 5 01009 Blount County, Alabama 21205
## 6 01011 Bullock County, Alabama 3429
## Margin.of.Error..Households..Total
## 1 366
## 2 1143
## 3 338
## 4 299
## 5 430
## 6 195
## Estimate..Households..Total..Less.than..10.000
## 1 6.2
## 2 5.2
## 3 14.6
## 4 11.0
## 5 10.1
## 6 18.2
## Margin.of.Error..Households..Total..Less.than..10.000
## 1 1.4
## 2 1.0
## 3 2.6
## 4 3.3
## 5 2.0
## 6 4.6
## Estimate..Households..Total...10.000.to..14.999
## 1 4.6
## 2 4.8
## 3 7.6
## 4 7.1
## 5 4.6
## 6 9.8
## Margin.of.Error..Households..Total...10.000.to..14.999
## 1 1.2
## 2 0.7
## 3 1.5
## 4 2.8
## 5 1.0
## 6 5.0
## Estimate..Households..Total...15.000.to..24.999
## 1 12.3
## 2 7.7
## 3 18.4
## 4 10.5
## 5 11.0
## 6 11.9
## Margin.of.Error..Households..Total...15.000.to..24.999
## 1 2.0
## 2 0.8
## 3 2.2
## 4 2.7
## 5 1.9
## 6 4.6
## Estimate..Households..Total...25.000.to..34.999
## 1 8.5
## 2 10.0
## 3 9.4
## 4 9.1
## 5 11.2
## 6 10.6
## Margin.of.Error..Households..Total...25.000.to..34.999
## 1 1.7
## 2 1.1
## 3 1.7
## 4 2.7
## 5 1.7
## 6 4.0
## Estimate..Households..Total...35.000.to..49.999
## 1 12.7
## 2 14.0
## 3 12.5
## 4 11.2
## 5 14.5
## 6 8.0
## Margin.of.Error..Households..Total...35.000.to..49.999
## 1 1.9
## 2 1.4
## 3 1.9
## 4 3.0
## 5 1.9
## 6 3.9
## Estimate..Households..Total...50.000.to..74.999
## 1 17.0
## 2 16.8
## 3 16.3
## 4 19.4
## 5 17.9
## 6 17.4
## Margin.of.Error..Households..Total...50.000.to..74.999
## 1 2.0
## 2 1.4
## 3 2.9
## 4 3.9
## 5 2.2
## 6 5.3
## Estimate..Households..Total...75.000.to..99.999
## 1 13.4
## 2 13.7
## 3 7.3
## 4 15.6
## 5 10.8
## 6 7.5
## Margin.of.Error..Households..Total...75.000.to..99.999
## 1 2.3
## 2 1.2
## 3 1.6
## 4 3.7
## 5 1.5
## 6 2.9
## Estimate..Households..Total...100.000.to..149.999
## 1 16.4
## 2 15.5
## 3 9.1
## 4 11.2
## 5 12.4
## 6 12.4
## Margin.of.Error..Households..Total...100.000.to..149.999
## 1 2.3
## 2 1.0
## 3 2.0
## 4 2.5
## 5 1.7
## 6 4.1
## Estimate..Households..Total...150.000.to..199.999
## 1 4.8
## 2 5.8
## 3 2.2
## 4 2.8
## 5 4.9
## 6 1.3
## Margin.of.Error..Households..Total...150.000.to..199.999
## 1 1.3
## 2 0.7
## 3 0.8
## 4 1.5
## 5 1.0
## 6 2.0
## Estimate..Households..Total...200.000.or.more
## 1 4.2
## 2 6.6
## 3 2.6
## 4 2.2
## 5 2.7
## 6 2.8
## Margin.of.Error..Households..Total...200.000.or.more
## 1 1.2
## 2 0.9
## 3 1.2
## 4 1.3
## 5 0.8
## 6 1.8
## Estimate..Households..Median.income..dollars.
## 1 57982
## 2 61756
## 3 34990
## 4 51721
## 5 48922
## 6 33866
## Margin.of.Error..Households..Median.income..dollars.
## 1 4839
## 2 2268
## 3 2909
## 4 6237
## 5 2269
## 6 10094
## Estimate..Households..Mean.income..dollars.
## 1 75614
## 2 83626
## 3 51557
## 4 61655
## 5 66360
## 6 50664
## Margin.of.Error..Households..Mean.income..dollars.
## 1 5718
## 2 2634
## 3 4229
## 4 5787
## 5 3864
## 6 5359We select columns Geography and
Estimate..Households..Median.income..dollars. and name a
new dataframe called USCensusIncome_select.
# Select columns and create a new dataframe called USCensusIncome_select
USCensusIncome_select <- USCensusIncome %>%
select(Geography, Estimate..Households..Median.income..dollars.)
head(USCensusIncome_select)## Geography Estimate..Households..Median.income..dollars.
## 1 01001 57982
## 2 01003 61756
## 3 01005 34990
## 4 01007 51721
## 5 01009 48922
## 6 01011 33866# Rename column in USCensusIncome_select
USCensusIncome_select <- USCensusIncome_select %>%
rename(Median.income = Estimate..Households..Median.income..dollars.)
head(USCensusIncome_select)## Geography Median.income
## 1 01001 57982
## 2 01003 61756
## 3 01005 34990
## 4 01007 51721
## 5 01009 48922
## 6 01011 33866Geography are both columns that match in
suicide_data and USCensusIncome_select. Let’s
join these dataframes.
# Join the dataframes on 'Geography'
suicide_data <- left_join(suicide_data, USCensusIncome_select, by = "Geography")
head(suicide_data)## County Geography Deaths Population Crude.Rate LND_SQMI
## 1 Autauga County, AL 01001 115 609875 18.900000 604.49
## 2 Baldwin County, AL 01003 444 2250866 19.700000 2027.08
## 3 Barbour County, AL 01005 41 287620 14.300000 904.59
## 4 Bibb County, AL 01007 38 248120 15.300000 625.50
## 5 Blount County, AL 01009 122 635351 19.200000 650.65
## 6 Bullock County, AL 01011 11 114915 9.572293 626.11
## population.density Median.income
## 1 1008.9083 57982
## 2 1110.3982 61756
## 3 317.9562 34990
## 4 396.6747 51721
## 5 976.4866 48922
## 6 183.5380 338663. percent of people who identify as white
USCensusDemographic has information on ethnic
identification by county.
head(USCensusDemographic)## Geography Geographic.Area.Name
## 1 0500000US01001 Autauga County, Alabama
## 2 0500000US01003 Baldwin County, Alabama
## 3 0500000US01005 Barbour County, Alabama
## 4 0500000US01007 Bibb County, Alabama
## 5 0500000US01009 Blount County, Alabama
## 6 0500000US01011 Bullock County, Alabama
## Estimate..SEX.AND.AGE..Total.population
## 1 55639
## 2 218289
## 3 25026
## 4 22374
## 5 57755
## 6 10173
## Estimate..RACE..Total.population..One.race
## 1 54218
## 2 212612
## 3 24504
## 4 22260
## 5 56462
## 6 10098
## Margin.of.Error..RACE..Total.population..One.race
## 1 397
## 2 1040
## 3 176
## 4 84
## 5 289
## 6 80
## Estimate..RACE..Total.population..Two.or.more.races
## 1 1421
## 2 5677
## 3 522
## 4 114
## 5 1293
## 6 75
## Margin.of.Error..RACE..Total.population..Two.or.more.races
## 1 397
## 2 1040
## 3 176
## 4 84
## 5 289
## 6 80
## Estimate..RACE..Total.population..One.race.1
## 1 54218
## 2 212612
## 3 24504
## 4 22260
## 5 56462
## 6 10098
## Margin.of.Error..RACE..Total.population..One.race.1
## 1 397
## 2 1040
## 3 176
## 4 84
## 5 289
## 6 80
## Estimate..RACE..Total.population..One.race..White
## 1 42150
## 2 186504
## 3 11587
## 4 17138
## 5 54271
## 6 2663
## Margin.of.Error..RACE..Total.population..One.race..White
## 1 207
## 2 1173
## 3 160
## 4 95
## 5 537
## 6 324
## Estimate..RACE..Total.population..One.race..Black.or.African.American
## 1 10866
## 2 19153
## 3 11929
## 4 5045
## 5 808
## 6 6980
## Margin.of.Error..RACE..Total.population..One.race..Black.or.African.American
## 1 347
## 2 728
## 3 183
## 4 142
## 5 173
## 6 140
## Estimate..RACE..Total.population..One.race..American.Indian.and.Alaska.Native
## 1 155
## 2 1514
## 3 88
## 4 12
## 5 55
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..American.Indian.and.Alaska.Native
## 1 102
## 2 418
## 3 65
## 4 25
## 5 47
## 6 19
## Estimate..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Cherokee.tribal.grouping
## 1 113
## 2 292
## 3 7
## 4 2
## 5 29
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Cherokee.tribal.grouping
## 1 101
## 2 233
## 3 13
## 4 17
## 5 31
## 6 19
## Margin.of.Error..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Chippewa.tribal.grouping
## 1 29
## 2 29
## 3 23
## 4 23
## 5 29
## 6 19
## Estimate..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Navajo.tribal.grouping
## 1 0
## 2 53
## 3 7
## 4 0
## 5 0
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Navajo.tribal.grouping
## 1 29
## 2 100
## 3 12
## 4 23
## 5 29
## 6 19
## Estimate..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Sioux.tribal.grouping
## 1 10
## 2 0
## 3 0
## 4 0
## 5 0
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..American.Indian.and.Alaska.Native..Sioux.tribal.grouping
## 1 17
## 2 29
## 3 23
## 4 23
## 5 29
## 6 19
## Estimate..RACE..Total.population..One.race..Asian
## 1 649
## 2 2033
## 3 122
## 4 56
## 5 236
## 6 137
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Asian
## 1 174
## 2 352
## 3 25
## 4 81
## 5 41
## 6 117
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Asian.Indian
## 1 11
## 2 469
## 3 65
## 4 56
## 5 65
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Asian..Asian.Indian
## 1 23
## 2 367
## 3 61
## 4 81
## 5 80
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Asian.Indian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Asian.Indian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Chinese
## 1 28
## 2 595
## 3 0
## 4 0
## 5 84
## 6 56
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Chinese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Asian..Chinese
## 1 58
## 2 358
## 3 23
## 4 23
## 5 67
## 6 99
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Chinese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Filipino
## 1 203
## 2 131
## 3 2
## 4 0
## 5 28
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Asian..Filipino
## 1 140
## 2 90
## 3 4
## 4 23
## 5 34
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Filipino
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Filipino
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Japanese
## 1 82
## 2 331
## 3 5
## 4 0
## 5 14
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Asian..Japanese
## 1 95
## 2 200
## 3 9
## 4 23
## 5 20
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Japanese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Japanese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Korean
## 1 200
## 2 112
## 3 14
## 4 0
## 5 45
## 6 0
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Korean
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Asian..Korean
## 1 220
## 2 133
## 3 28
## 4 23
## 5 56
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Korean
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Vietnamese
## 1 112
## 2 173
## 3 15
## 4 0
## 5 0
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Asian..Vietnamese
## 1 160
## 2 178
## 3 22
## 4 23
## 5 29
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Vietnamese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Vietnamese
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Asian..Other.Asian
## 1 13
## 2 222
## 3 21
## 4 0
## 5 0
## 6 81
## Margin.of.Error..RACE..Total.population..One.race..Asian..Other.Asian
## 1 22
## 2 204
## 3 42
## 4 23
## 5 29
## 6 70
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Asian..Other.Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Asian..Other.Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander
## 1 24
## 2 10
## 3 1
## 4 0
## 5 55
## 6 0
## Annotation.of.Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander
## 1 37
## 2 16
## 3 2
## 4 23
## 5 60
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Native.Hawaiian
## 1 0
## 2 0
## 3 1
## 4 0
## 5 28
## 6 0
## Annotation.of.Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Native.Hawaiian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Native.Hawaiian
## 1 29
## 2 29
## 3 2
## 4 23
## 5 48
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Native.Hawaiian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Chamorro
## 1 24
## 2 10
## 3 0
## 4 0
## 5 27
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Chamorro
## 1 37
## 2 16
## 3 23
## 4 23
## 5 42
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Chamorro
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Chamorro
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Samoan
## 1 0
## 2 0
## 3 0
## 4 0
## 5 0
## 6 0
## Annotation.of.Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Samoan
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Samoan
## 1 29
## 2 29
## 3 23
## 4 23
## 5 29
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Samoan
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Other.Pacific.Islander
## 1 0
## 2 0
## 3 0
## 4 0
## 5 0
## 6 0
## Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Other.Pacific.Islander
## 1 29
## 2 29
## 3 23
## 4 23
## 5 29
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Other.Pacific.Islander
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Native.Hawaiian.and.Other.Pacific.Islander..Other.Pacific.Islander
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..One.race..Some.other.race
## 1 374
## 2 3398
## 3 777
## 4 9
## 5 1037
## 6 318
## Margin.of.Error..RACE..Total.population..One.race..Some.other.race
## 1 283
## 2 926
## 3 218
## 4 19
## 5 499
## 6 324
## Annotation.of.Margin.of.Error..RACE..Total.population..One.race..Some.other.race
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..One.race..Some.other.race
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..Two.or.more.races.1
## 1 1421
## 2 5677
## 3 522
## 4 114
## 5 1293
## 6 75
## Annotation.of.Estimate..RACE..Total.population..Two.or.more.races
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..Two.or.more.races.1
## 1 397
## 2 1040
## 3 176
## 4 84
## 5 289
## 6 80
## Annotation.of.Margin.of.Error..RACE..Total.population..Two.or.more.races
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..Two.or.more.races..White.and.Black.or.African.American
## 1 384
## 2 733
## 3 26
## 4 25
## 5 227
## 6 0
## Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.Black.or.African.American
## 1 243
## 2 357
## 3 26
## 4 26
## 5 146
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.Black.or.African.American
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..Two.or.more.races..White.and.Black.or.African.American
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..Two.or.more.races..White.and.American.Indian.and.Alaska.Native
## 1 274
## 2 1589
## 3 76
## 4 78
## 5 654
## 6 13
## Annotation.of.Estimate..RACE..Total.population..Two.or.more.races..White.and.American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.American.Indian.and.Alaska.Native
## 1 102
## 2 399
## 3 65
## 4 79
## 5 159
## 6 27
## Annotation.of.Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..Two.or.more.races..White.and.Asian
## 1 244
## 2 943
## 3 10
## 4 2
## 5 33
## 6 0
## Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.Asian
## 1 174
## 2 354
## 3 19
## 4 4
## 5 41
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..Two.or.more.races..White.and.Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..Two.or.more.races..White.and.Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..RACE..Total.population..Two.or.more.races..Black.or.African.American.and.American.Indian.and.Alaska.Native
## 1 18
## 2 77
## 3 95
## 4 0
## 5 0
## 6 0
## Margin.of.Error..RACE..Total.population..Two.or.more.races..Black.or.African.American.and.American.Indian.and.Alaska.Native
## 1 25
## 2 97
## 3 84
## 4 23
## 5 29
## 6 19
## Annotation.of.Margin.of.Error..RACE..Total.population..Two.or.more.races..Black.or.African.American.and.American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..RACE..Total.population..Two.or.more.races..Black.or.African.American.and.American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population
## 1 55639
## 2 218289
## 3 25026
## 4 22374
## 5 57755
## 6 10173
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population
## 1 *****
## 2 *****
## 3 *****
## 4 *****
## 5 *****
## 6 *****
## Annotation.of.Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population
## 1 *****
## 2 *****
## 3 *****
## 4 *****
## 5 *****
## 6 *****
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..White
## 1 43422
## 2 191923
## 3 11989
## 4 17243
## 5 55555
## 6 2683
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..White
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..White
## 1 396
## 2 1392
## 3 232
## 4 104
## 5 511
## 6 324
## Annotation.of.Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..White
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Black.or.African.American
## 1 11423
## 2 20377
## 3 12140
## 4 5079
## 5 1098
## 6 7035
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Black.or.African.American
## 1 129
## 2 550
## 3 124
## 4 142
## 5 77
## 6 121
## Annotation.of.Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Black.or.African.American
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Black.or.African.American
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..American.Indian.and.Alaska.Native
## 1 447
## 2 3347
## 3 315
## 4 90
## 5 796
## 6 13
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..American.Indian.and.Alaska.Native
## 1 25
## 2 341
## 3 158
## 4 86
## 5 149
## 6 27
## Annotation.of.Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..American.Indian.and.Alaska.Native
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Asian
## 1 1048
## 2 3120
## 3 155
## 4 67
## 5 288
## 6 192
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Asian
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Asian
## 1 140
## 2 145
## 3 25
## 4 82
## 5 27
## 6 138
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Native.Hawaiian.and.Other.Pacific.Islander
## 1 24
## 2 570
## 3 28
## 4 0
## 5 83
## 6 0
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Native.Hawaiian.and.Other.Pacific.Islander
## 1 37
## 2 537
## 3 32
## 4 23
## 5 67
## 6 19
## Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Some.other.race
## 1 720
## 2 4928
## 3 1009
## 4 9
## 5 1315
## 6 325
## Annotation.of.Estimate..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Some.other.race
## 1 null
## 2 null
## 3 null
## 4 null
## 5 null
## 6 null
## Margin.of.Error..Race.alone.or.in.combination.with.one.or.more.other.races..Total.population..Some.other.race
## 1 251
## 2 1102
## 3 194
## 4 19
## 5 525
## 6 324
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population
## 1 55639
## 2 218289
## 3 25026
## 4 22374
## 5 57755
## 6 10173
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race.
## 1 1601
## 2 9947
## 3 1110
## 4 600
## 5 5362
## 6 824
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Mexican
## 1 742
## 2 5206
## 3 844
## 4 162
## 5 4777
## 6 756
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Mexican
## 1 259
## 2 915
## 3 159
## 4 92
## 5 306
## 6 126
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Puerto.Rican
## 1 423
## 2 1217
## 3 74
## 4 61
## 5 128
## 6 67
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Puerto.Rican
## 1 229
## 2 608
## 3 44
## 4 83
## 5 111
## 6 126
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Cuban
## 1 211
## 2 423
## 3 0
## 4 19
## 5 68
## 6 0
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Cuban
## 1 204
## 2 228
## 3 23
## 4 25
## 5 101
## 6 19
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Other.Hispanic.or.Latino
## 1 225
## 2 3101
## 3 192
## 4 358
## 5 389
## 6 1
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Hispanic.or.Latino..of.any.race...Other.Hispanic.or.Latino
## 1 165
## 2 842
## 3 158
## 4 113
## 5 269
## 6 3
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino
## 1 54038
## 2 208342
## 3 23916
## 4 21774
## 5 52393
## 6 9349
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..White.alone
## 1 41160
## 2 180955
## 3 11332
## 4 16650
## 5 50065
## 6 2162
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..White.alone
## 1 78
## 2 654
## 3 83
## 4 23
## 5 250
## 6 19
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Black.or.African.American.alone
## 1 10849
## 2 19027
## 3 11889
## 4 4971
## 5 771
## 6 6980
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Black.or.African.American.alone
## 1 345
## 2 744
## 3 179
## 4 111
## 5 167
## 6 140
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..American.Indian.and.Alaska.Native.alone
## 1 155
## 2 1327
## 3 81
## 4 12
## 5 49
## 6 0
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..American.Indian.and.Alaska.Native.alone
## 1 102
## 2 371
## 3 63
## 4 25
## 5 41
## 6 19
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Asian.alone
## 1 649
## 2 2033
## 3 122
## 4 56
## 5 236
## 6 137
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Asian.alone
## 1 174
## 2 352
## 3 25
## 4 81
## 5 41
## 6 117
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Native.Hawaiian.and.Other.Pacific.Islander.alone
## 1 0
## 2 10
## 3 1
## 4 0
## 5 55
## 6 0
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Native.Hawaiian.and.Other.Pacific.Islander.alone
## 1 29
## 2 16
## 3 2
## 4 23
## 5 60
## 6 19
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Some.other.race.alone
## 1 101
## 2 740
## 3 157
## 4 0
## 5 179
## 6 2
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Some.other.race.alone
## 1 145
## 2 534
## 3 110
## 4 23
## 5 225
## 6 4
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races
## 1 1124
## 2 4250
## 3 334
## 4 85
## 5 1038
## 6 68
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races
## 1 374
## 2 871
## 3 149
## 4 67
## 5 181
## 6 76
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races..Two.races.including.Some.other.race
## 1 54
## 2 299
## 3 44
## 4 0
## 5 107
## 6 0
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races..Two.races.including.Some.other.race
## 1 57
## 2 231
## 3 47
## 4 23
## 5 95
## 6 19
## Estimate..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races..Two.races.excluding.Some.other.race..and.Three.or.more.races
## 1 1070
## 2 3951
## 3 290
## 4 85
## 5 931
## 6 68
## Margin.of.Error..HISPANIC.OR.LATINO.AND.RACE..Total.population..Not.Hispanic.or.Latino..Two.or.more.races..Two.races.excluding.Some.other.race..and.Three.or.more.races
## 1 374
## 2 854
## 3 142
## 4 67
## 5 160
## 6 76We need to calculate the percentage of white by dividing
Estimate..RACE..Total population..One race..White by
Estimate..SEX.AND.AGE..Total.population
USCensusDemographic <- USCensusDemographic %>%
mutate(percent.white = (Estimate..RACE..Total.population..One.race..White / Estimate..SEX.AND.AGE..Total.population) * 100)We select the column
Estimate..RACE..Total population..One race..White and
Geography.
# Select columns and create a new dataframe called USCensusDemographic_select
USCensusDemographic_select <- USCensusDemographic %>%
select(Geography, percent.white)
head(USCensusDemographic_select)## Geography percent.white
## 1 0500000US01001 75.75621
## 2 0500000US01003 85.43903
## 3 0500000US01005 46.29985
## 4 0500000US01007 76.59784
## 5 0500000US01009 93.96762
## 6 0500000US01011 26.17714We remove the “0500000US” part in the Geography column for the USCensusDemographic_select dataframe.
# Remove "0500000US" from the 'Geography' column in USCensusDemographic_select
USCensusDemographic_select$Geography <- gsub("0500000US", "", USCensusDemographic_select$Geography)
head(USCensusDemographic_select)## Geography percent.white
## 1 01001 75.75621
## 2 01003 85.43903
## 3 01005 46.29985
## 4 01007 76.59784
## 5 01009 93.96762
## 6 01011 26.17714We use the left_join() function to join
USCensusDemographic_select to suicide_data by
the Geography column:
# Join USCensusDemographic_select to suicide_data
suicide_data <- suicide_data %>%
left_join(USCensusDemographic_select, by = "Geography")
head(suicide_data)## County Geography Deaths Population Crude.Rate LND_SQMI
## 1 Autauga County, AL 01001 115 609875 18.900000 604.49
## 2 Baldwin County, AL 01003 444 2250866 19.700000 2027.08
## 3 Barbour County, AL 01005 41 287620 14.300000 904.59
## 4 Bibb County, AL 01007 38 248120 15.300000 625.50
## 5 Blount County, AL 01009 122 635351 19.200000 650.65
## 6 Bullock County, AL 01011 11 114915 9.572293 626.11
## population.density Median.income percent.white
## 1 1008.9083 57982 75.75621
## 2 1110.3982 61756 85.43903
## 3 317.9562 34990 46.29985
## 4 396.6747 51721 76.59784
## 5 976.4866 48922 93.96762
## 6 183.5380 33866 26.177144. percentage of people who live alone
USCensusHousehold has information on household formation
using data from the American Community Survey. We select the column
Estimate..Total..Total.households..SELECTED.HOUSEHOLDS.BY.TYPE..Householder.living.alone,
which provides a 5-year estimate of the percentage of people living
alone in each county, and Geography.
# Select columns and create a new dataframe called USCensusHousehold_select
USCensusHousehold_select <- USCensusHousehold %>%
select(Geography, Estimate..Total..Total.households..SELECTED.HOUSEHOLDS.BY.TYPE..Householder.living.alone)
head(USCensusHousehold_select)## Geography
## 1 0500000US01001
## 2 0500000US01003
## 3 0500000US01005
## 4 0500000US01007
## 5 0500000US01009
## 6 0500000US01011
## Estimate..Total..Total.households..SELECTED.HOUSEHOLDS.BY.TYPE..Householder.living.alone
## 1 26.3
## 2 29.0
## 3 32.8
## 4 25.9
## 5 25.9
## 6 35.4We remove “0500000US” from the Geography column in USCensusHousehold_select.
# Remove "0500000US" from the 'Geography' column in USCensusHousehold_select
USCensusHousehold_select$Geography <- gsub("0500000US", "", USCensusHousehold_select$Geography)
head(USCensusHousehold_select)## Geography
## 1 01001
## 2 01003
## 3 01005
## 4 01007
## 5 01009
## 6 01011
## Estimate..Total..Total.households..SELECTED.HOUSEHOLDS.BY.TYPE..Householder.living.alone
## 1 26.3
## 2 29.0
## 3 32.8
## 4 25.9
## 5 25.9
## 6 35.4Let’s rename the columns in USCensusHousehold_select
# Rename column in USCensusHousehold_select
USCensusHousehold_select <- USCensusHousehold_select %>%
rename(percentage.alone = Estimate..Total..Total.households..SELECTED.HOUSEHOLDS.BY.TYPE..Householder.living.alone)
head(USCensusHousehold_select)## Geography percentage.alone
## 1 01001 26.3
## 2 01003 29.0
## 3 01005 32.8
## 4 01007 25.9
## 5 01009 25.9
## 6 01011 35.4Let’s join USCensusHousehold_select to
suicide_data using the Geography column.
# Join USCensusHousehold_select to suicide_data
suicide_data <- suicide_data %>%
left_join(USCensusHousehold_select, by = "Geography")
head(suicide_data)## County Geography Deaths Population Crude.Rate LND_SQMI
## 1 Autauga County, AL 01001 115 609875 18.900000 604.49
## 2 Baldwin County, AL 01003 444 2250866 19.700000 2027.08
## 3 Barbour County, AL 01005 41 287620 14.300000 904.59
## 4 Bibb County, AL 01007 38 248120 15.300000 625.50
## 5 Blount County, AL 01009 122 635351 19.200000 650.65
## 6 Bullock County, AL 01011 11 114915 9.572293 626.11
## population.density Median.income percent.white percentage.alone
## 1 1008.9083 57982 75.75621 26.3
## 2 1110.3982 61756 85.43903 29.0
## 3 317.9562 34990 46.29985 32.8
## 4 396.6747 51721 76.59784 25.9
## 5 976.4866 48922 93.96762 25.9
## 6 183.5380 33866 26.17714 35.4Recoding Data
Let’s select the important columns from suicide_data
# Select important columns from suicide_data
suicide_data_select <- suicide_data %>%
select(Geography, Deaths, Crude.Rate, population.density, Median.income, percent.white, percentage.alone)
head(suicide_data_select)## Geography Deaths Crude.Rate population.density Median.income percent.white
## 1 01001 115 18.900000 1008.9083 57982 75.75621
## 2 01003 444 19.700000 1110.3982 61756 85.43903
## 3 01005 41 14.300000 317.9562 34990 46.29985
## 4 01007 38 15.300000 396.6747 51721 76.59784
## 5 01009 122 19.200000 976.4866 48922 93.96762
## 6 01011 11 9.572293 183.5380 33866 26.17714
## percentage.alone
## 1 26.3
## 2 29.0
## 3 32.8
## 4 25.9
## 5 25.9
## 6 35.4We need to make sure the relevant columns are in the correct class for analysis. Let’s convert the necessary columns to the appropriate class.
# Check the class of each column
sapply(suicide_data_select, class)## Geography Deaths Crude.Rate population.density
## "character" "character" "numeric" "numeric"
## Median.income percent.white percentage.alone
## "character" "numeric" "numeric"# Convert necessary columns to appropriate class
suicide_data_select$Deaths <- as.numeric(as.character(suicide_data_select$Deaths))## Warning: NAs introduced by coercionsuicide_data_select$Crude.Rate <- as.numeric(as.character(suicide_data_select$Crude.Rate))
suicide_data_select$Median.income <- as.numeric(as.character(suicide_data_select$Median.income))## Warning: NAs introduced by coercionsuicide_data_select$percent.white <- as.numeric(suicide_data_select$percent.white)
# Check the class of each column again
sapply(suicide_data_select, class)## Geography Deaths Crude.Rate population.density
## "character" "numeric" "numeric" "numeric"
## Median.income percent.white percentage.alone
## "numeric" "numeric" "numeric"# Check for missing values
colSums(is.na(suicide_data_select[c("Crude.Rate", "population.density", "Median.income", "percent.white", "percentage.alone")]))## Crude.Rate population.density Median.income percent.white
## 334 10 10 9
## percentage.alone
## 9Before proceeding with analysis, we would address missing values. Depending on the context and the amount of missing data, we’d consider different strategies such as imputation, using complete cases only, or exploring the reasons behind the missing data to understand any potential biases.
Analyses
Step 3: Descriptive Statistics
Now that we have all the files merged, we conduct descriptive
statistics on each of the variables of interest
(Crude.Rate, population.density (the number of
people / sq mile of land area), Median.income,
percent.white, percentage.alone)
To make sure we don’t have any values that need to be recoded. Whenever we are working with income data, we should see if it is normally distributed or would be better represented as a log transformation.
# Crude.Rate
hist(suicide_data_select$Crude.Rate)
hist(log10(suicide_data_select$Crude.Rate + 1))
This data appears normal.
# population.density
hist(suicide_data_select$population.density)
hist(log10(suicide_data_select$population.density + 1))
This data appears normal.
# Median.income
hist(suicide_data_select$Median.income)
hist(log10(suicide_data_select$Median.income + 1))
This data appears normal.
# percent.white
hist(suicide_data_select$percent.white)
hist(log10(suicide_data_select$percent.white + 1))
The percent white is skewed.
# percentage.alone
hist(suicide_data_select$percentage.alone)
hist(log10(suicide_data_select$percentage.alone + 1))
This data appears normal.
# Descriptive statistics for variables of interest
summary(suicide_data_select[c("Crude.Rate", "population.density", "Median.income", "percent.white", "percentage.alone")])## Crude.Rate population.density Median.income percent.white
## Min. : 5.20 Min. : 0.4 Min. : 22292 Min. : 9.352
## 1st Qu.: 13.60 1st Qu.: 176.4 1st Qu.: 45668 1st Qu.: 74.681
## Median : 16.80 Median : 469.8 Median : 52856 Median : 88.102
## Mean : 17.74 Mean : 2460.4 Mean : 55026 Mean : 81.814
## 3rd Qu.: 20.50 3rd Qu.: 1199.9 3rd Qu.: 61502 3rd Qu.: 94.087
## Max. :115.60 Max. :531226.6 Max. :147111 Max. :100.000
## NA's :334 NA's :10 NA's :10 NA's :9
## percentage.alone
## Min. : 8.50
## 1st Qu.:25.60
## Median :28.80
## Mean :28.79
## 3rd Qu.:31.70
## Max. :71.00
## NA's :9Stepwise Model
Step 4: Let’s use a stepwise model to determine which variables should be included in our “best” model. Run at least 2 of the following 3 models: backwards stepwise, forward stepwise, or both.
# Remove rows with NA values from the dataset
suicide_data_select_omit <- na.omit(suicide_data_select)# Fit the full model using the cleaned data
fullmodel <- lm(Crude.Rate ~ population.density + Median.income +
percent.white + percentage.alone, data = suicide_data_select_omit)
# Perform backward stepwise selection
step(fullmodel, direction="backward")## Start: AIC=10014.45
## Crude.Rate ~ population.density + Median.income + percent.white +
## percentage.alone
##
## Df Sum of Sq RSS AIC
## <none> 98928 10014
## - percentage.alone 1 98.5 99027 10015
## - population.density 1 960.0 99889 10040
## - percent.white 1 3350.6 102279 10106
## - Median.income 1 3966.3 102895 10123##
## Call:
## lm(formula = Crude.Rate ~ population.density + Median.income +
## percent.white + percentage.alone, data = suicide_data_select_omit)
##
## Coefficients:
## (Intercept) population.density Median.income percent.white
## 1.604e+01 -4.213e-05 -9.228e-05 6.860e-02
## percentage.alone
## 4.523e-02Based on the stepwise model selection output, the model with no variables removed has the lowest AIC (10014).
#forward
full_model <- formula(lm(Crude.Rate ~ population.density + Median.income +
percent.white + percentage.alone, data=(suicide_data_select_omit)))
intercept_only <- lm(Crude.Rate ~ 1, data=na.omit(suicide_data_select_omit))
step(intercept_only, direction="forward", scope=full_model)## Start: AIC=10313.5
## Crude.Rate ~ 1
##
## Df Sum of Sq RSS AIC
## + Median.income 1 6133.1 104222 10155
## + percent.white 1 3370.8 106985 10228
## + population.density 1 3241.9 107114 10232
## + percentage.alone 1 1224.5 109131 10284
## <none> 110355 10314
##
## Step: AIC=10154.88
## Crude.Rate ~ Median.income
##
## Df Sum of Sq RSS AIC
## + percent.white 1 4316.1 99906 10038
## + population.density 1 1856.7 102366 10106
## <none> 104222 10155
## + percentage.alone 1 0.7 104222 10157
##
## Step: AIC=10038.07
## Crude.Rate ~ Median.income + percent.white
##
## Df Sum of Sq RSS AIC
## + population.density 1 879.3 99027 10015
## <none> 99906 10038
## + percentage.alone 1 17.8 99889 10040
##
## Step: AIC=10015.24
## Crude.Rate ~ Median.income + percent.white + population.density
##
## Df Sum of Sq RSS AIC
## + percentage.alone 1 98.502 98928 10014
## <none> 99027 10015
##
## Step: AIC=10014.45
## Crude.Rate ~ Median.income + percent.white + population.density +
## percentage.alone##
## Call:
## lm(formula = Crude.Rate ~ Median.income + percent.white + population.density +
## percentage.alone, data = na.omit(suicide_data_select_omit))
##
## Coefficients:
## (Intercept) Median.income percent.white population.density
## 1.604e+01 -9.228e-05 6.860e-02 -4.213e-05
## percentage.alone
## 4.523e-02Based on the forward stepwise model selection output, the final model includes all four variables. These results are consistent with those obtained from the backward stepwise selection.
# Fit the full model using the cleaned data
fullmodel <- lm(Crude.Rate ~ population.density + Median.income +
percent.white + percentage.alone, data = suicide_data_select_omit)
# Perform stepwise selection (both forward and backward)
step(fullmodel, direction="both")## Start: AIC=10014.45
## Crude.Rate ~ population.density + Median.income + percent.white +
## percentage.alone
##
## Df Sum of Sq RSS AIC
## <none> 98928 10014
## - percentage.alone 1 98.5 99027 10015
## - population.density 1 960.0 99889 10040
## - percent.white 1 3350.6 102279 10106
## - Median.income 1 3966.3 102895 10123##
## Call:
## lm(formula = Crude.Rate ~ population.density + Median.income +
## percent.white + percentage.alone, data = suicide_data_select_omit)
##
## Coefficients:
## (Intercept) population.density Median.income percent.white
## 1.604e+01 -4.213e-05 -9.228e-05 6.860e-02
## percentage.alone
## 4.523e-02The stepwise selection process found that the best model includes all four predictor variables since removing any of them would result in an increased AIC value.
Step 5: Interprate Findings
Based on the analysis of the county-level data, we found that four predictor variables had a significant association with the Crude Rate of suicide.
The results show a negative association between population density and the Crude Rate of suicide. This implies that, on average, counties with higher population density tend to have lower suicide rates, while controlling for other factors in the model.
There is a negative association between median income and the Crude Rate of suicide. This suggests that, on average, counties with higher median income levels tend to have lower suicide rates, while controlling for other factors in the model.
The results show a positive association between the percentage of the white population in a county and the Crude Rate of suicide. This indicates that, on average, counties with a higher proportion of white residents tend to have higher suicide rates, while controlling for other factors in the model.
There is a positive association between the percentage of people living alone in a county and the Crude Rate of suicide. This suggests that, on average, counties with a higher proportion of individuals living alone tend to have higher suicide rates, while controlling for other factors in the model.