Story 6

Overview

The United Nations Food and Agriculture Organization publication, The State of Food Security and Nutrition in the World 2022 (https://www.fao.org/documents/card/en/c/cc0639en) might lead one to the conclusion that it’s an elsewhere problem. That the people who are suffering malnutrition and starvation are “elsewhere”, not in our backyard. For this assignment you will need to take a closer look here at home (the US)

Notes:

You will need to locate and source data that reflects food security and nutrition by state broken down by men, women, children and by age groups

Your analysis should demonstrate correlations that exist between level of poverty and food insecurity, malnutrition and starvation.

Your data and analysis should also indicate what happens to the children as they mature into adults. Will they become fully functional citizens or will they require continued support?

You data visualizations need to tell the story for a political audience that you were lobbying to address the issue of food insecurity in the US

Loading datasets

Loading household dataset

#Loading data from CSV file
data <- read.csv("C:/Users/aleja/Desktop/food-data.csv")

#View of the first few rows of dataset
head(data)

##   Year              Category              Subcategory         Sub.subcategory
## 1 2001        All households                                                 
## 2 2001 Household composition With children < 18 years                        
## 3 2001 Household composition With children < 18 years With children < 6 years
## 4 2001 Household composition With children < 18 years Married-couple families
## 5 2001 Household composition With children < 18 years  Female head, no spouse
## 6 2001 Household composition With children < 18 years    Male head, no spouse
##    Total Food.secure.1.000 Food.secure.percent Food.insecure.1.000
## 1 107824             96303                89.3               11521
## 2  38330             32141                83.9                6189
## 3  16858             13920                82.6                2938
## 4  26182             23389                89.3                2793
## 5   9080              6185                68.1                2895
## 6   2389              2009                84.1                 380
##   Food.insecure.percent Low.food.security.1.000 Low.food.security.percent
## 1                  10.7                    8010                       7.4
## 2                  16.1                    4744                      12.4
## 3                  17.4                    2304                      13.7
## 4                  10.7                    2247                       8.6
## 5                  31.9                    2101                      23.1
## 6                  15.9                     298                      12.5
##   Very.low.food.security.1.000 Very.low.food.security.percent
## 1                         3511                            3.3
## 2                         1445                            3.8
## 3                          634                            3.8
## 4                          546                            2.1
## 5                          794                            8.7
## 6                           82                            3.4

#dataset
str(data)

## 'data.frame':    660 obs. of  13 variables:
##  $ Year                          : int  2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 ...
##  $ Category                      : chr  "All households" "Household composition" "Household composition" "Household composition" ...
##  $ Subcategory                   : chr  "" "With children < 18 years" "With children < 18 years" "With children < 18 years" ...
##  $ Sub.subcategory               : chr  "" "" "With children < 6 years" "Married-couple families" ...
##  $ Total                         : int  107824 38330 16858 26182 9080 2389 678 69495 40791 16513 ...
##  $ Food.secure.1.000             : int  96303 32141 13920 23389 6185 2009 555 64163 38328 14915 ...
##  $ Food.secure.percent           : num  89.3 83.9 82.6 89.3 68.1 84.1 81.9 92.3 94 90.3 ...
##  $ Food.insecure.1.000           : int  11521 6189 2938 2793 2895 380 123 5332 2463 1598 ...
##  $ Food.insecure.percent         : num  10.7 16.1 17.4 10.7 31.9 15.9 18.1 7.7 6 9.7 ...
##  $ Low.food.security.1.000       : chr  "8010" "4744" "2304" "2247" ...
##  $ Low.food.security.percent     : num  7.4 12.4 13.7 8.6 23.1 12.5 14.6 4.7 3.9 5.8 ...
##  $ Very.low.food.security.1.000  : chr  "3511" "1445" "634" "546" ...
##  $ Very.low.food.security.percent: num  3.3 3.8 3.8 2.1 8.7 3.4 3.5 3 2.1 3.9 ...

Loading state dataset

#Loading data from CSV file
state <- read.csv("C:/Users/aleja/Desktop/food by state.csv")

#View of the first few rows of dataset
head(state)

##        Year      State Food.insecurity.prevalence
## 1 2006–2008 U.S. total                       12.2
## 2 2006–2008         AK                       11.6
## 3 2006–2008         AL                       13.3
## 4 2006–2008         AR                       15.9
## 5 2006–2008         AZ                       13.2
## 6 2006–2008         CA                       12.0
##   Food.insecurity.margin.of.error Very.low.food.security.prevalence
## 1                            0.25                               4.6
## 2                            1.66                               4.4
## 3                            1.66                               5.4
## 4                            3.19                               5.6
## 5                            1.51                               4.9
## 6                            0.74                               4.3
##   Very.low.food.security.margin.of.error
## 1                                   0.18
## 2                                   1.31
## 3                                   1.02
## 4                                   1.50
## 5                                   0.84
## 6                                   0.48

#dataset
str(state)

## 'data.frame':    780 obs. of  6 variables:
##  $ Year                                  : chr  "2006–2008" "2006–2008" "2006–2008" "2006–2008" ...
##  $ State                                 : chr  "U.S. total" "AK" "AL" "AR" ...
##  $ Food.insecurity.prevalence            : num  12.2 11.6 13.3 15.9 13.2 12 11.6 11 12.4 9.4 ...
##  $ Food.insecurity.margin.of.error       : num  0.25 1.66 1.66 3.19 1.51 0.74 1.13 1.53 1.15 0.98 ...
##  $ Very.low.food.security.prevalence     : num  4.6 4.4 5.4 5.6 4.9 4.3 5 4.1 4.2 3.7 ...
##  $ Very.low.food.security.margin.of.error: num  0.18 1.31 1.02 1.5 0.84 0.48 0.67 1.07 0.73 0.73 ...

Data preparation

For household dataset

#Looking for missing values
missing_values <- sapply(data, function(x) sum(is.na(x)))
print(missing_values)

##                           Year                       Category 
##                              0                              0 
##                    Subcategory                Sub.subcategory 
##                              0                              0 
##                          Total              Food.secure.1.000 
##                              0                              0 
##            Food.secure.percent            Food.insecure.1.000 
##                              0                              0 
##          Food.insecure.percent        Low.food.security.1.000 
##                              0                              6 
##      Low.food.security.percent   Very.low.food.security.1.000 
##                              7                              5 
## Very.low.food.security.percent 
##                              7

#Dropping rows with missing values
cleaned_data <- na.omit(data)

#Checking the dimensions of the cleaned dataset
dim(cleaned_data)

## [1] 653  13

#Converting Subcategory and Sub.subcategory to lowercase for case-insensitive matching
cleaned_data$Subcategory <- tolower(cleaned_data$Subcategory)
cleaned_data$Sub.subcategory <- tolower(cleaned_data$Sub.subcategory)

#Function to determine gender based on subcategory
get_gender <- function(subcategory) {
  if (grepl("female|woman|girl|women", subcategory)) {
    return("F")
  } else if (grepl("male|man|boy|men", subcategory)) {
    return("M")
  } else {
    return("Unknown")
  }
}

#function to create Gender column
cleaned_data$Gender <- mapply(get_gender, cleaned_data$Subcategory)

#function to Sub.subcategory if Gender is still unknown
unknown_indices <- which(cleaned_data$Gender == "Unknown")
cleaned_data$Gender[unknown_indices] <- mapply(get_gender, cleaned_data$Sub.subcategory[unknown_indices])

#Replacing empty strings with "Unknown"
cleaned_data$Gender[cleaned_data$Gender == ""] <- "Unknown"

#Printing the first few rows to verify the Gender column
head(cleaned_data)

##   Year              Category              Subcategory         Sub.subcategory
## 1 2001        All households                                                 
## 2 2001 Household composition with children < 18 years                        
## 3 2001 Household composition with children < 18 years with children < 6 years
## 4 2001 Household composition with children < 18 years married-couple families
## 5 2001 Household composition with children < 18 years  female head, no spouse
## 6 2001 Household composition with children < 18 years    male head, no spouse
##    Total Food.secure.1.000 Food.secure.percent Food.insecure.1.000
## 1 107824             96303                89.3               11521
## 2  38330             32141                83.9                6189
## 3  16858             13920                82.6                2938
## 4  26182             23389                89.3                2793
## 5   9080              6185                68.1                2895
## 6   2389              2009                84.1                 380
##   Food.insecure.percent Low.food.security.1.000 Low.food.security.percent
## 1                  10.7                    8010                       7.4
## 2                  16.1                    4744                      12.4
## 3                  17.4                    2304                      13.7
## 4                  10.7                    2247                       8.6
## 5                  31.9                    2101                      23.1
## 6                  15.9                     298                      12.5
##   Very.low.food.security.1.000 Very.low.food.security.percent  Gender
## 1                         3511                            3.3 Unknown
## 2                         1445                            3.8 Unknown
## 3                          634                            3.8 Unknown
## 4                          546                            2.1 Unknown
## 5                          794                            8.7       F
## 6                           82                            3.4       M

summary(cleaned_data)

##       Year        Category         Subcategory        Sub.subcategory   
##  Min.   :2001   Length:653         Length:653         Length:653        
##  1st Qu.:2006   Class :character   Class :character   Class :character  
##  Median :2011   Mode  :character   Mode  :character   Mode  :character  
##  Mean   :2011                                                           
##  3rd Qu.:2017                                                           
##  Max.   :2022                                                           
##      Total        Food.secure.1.000 Food.secure.percent Food.insecure.1.000
##  Min.   :   496   Min.   :   366    Min.   :57.00       Min.   :  105      
##  1st Qu.: 15613   1st Qu.: 12000    1st Qu.:81.50       1st Qu.: 2426      
##  Median : 25180   Median : 20717    Median :87.30       Median : 3447      
##  Mean   : 34568   Mean   : 30091    Mean   :84.26       Mean   : 4476      
##  3rd Qu.: 43842   3rd Qu.: 38929    3rd Qu.:90.20       3rd Qu.: 5944      
##  Max.   :132730   Max.   :118533    Max.   :95.10       Max.   :17853      
##  Food.insecure.percent Low.food.security.1.000 Low.food.security.percent
##  Min.   : 4.90         Length:653              Min.   : 3.200           
##  1st Qu.: 9.80         Class :character        1st Qu.: 5.900           
##  Median :12.70         Mode  :character        Median : 7.800           
##  Mean   :15.74                                 Mean   : 9.984           
##  3rd Qu.:18.50                                 3rd Qu.:13.200           
##  Max.   :43.00                                 Max.   :25.300           
##  Very.low.food.security.1.000 Very.low.food.security.percent    Gender         
##  Length:653                   Min.   : 0.800                 Length:653        
##  Class :character             1st Qu.: 3.600                 Class :character  
##  Mode  :character             Median : 4.700                 Mode  :character  
##                               Mean   : 5.752                                   
##                               3rd Qu.: 6.500                                   
##                               Max.   :19.300

For State dataset

#Looking for missing values
na_count <- sum(is.na(state))
cat("Number of missing values:", na_count, "\n")

## Number of missing values: 0

No missing values.

Creating regions for the analysis

#Defining the regions
northeast_states <- c("CT", "ME", "MA", "NH", "RI", "VT", "NJ", "NY", "PA", "DC")
midwest_states <- c("IL", "IN", "IA", "KS", "MI", "MN", "MO", "NE", "ND", "OH", "SD", "WI")
south_states <- c("AL", "AR", "DE", "FL", "GA", "KY", "LA", "MD", "MS", "NC", "OK", "SC", "TN", "TX", "VA", "WV")
west_states <- c("AK", "AZ", "CA", "CO", "HI", "ID", "MT", "NV", "NM", "OR", "UT", "WA", "WY")

#Assigning states to regions
state$Region <- ifelse(state$State %in% northeast_states, "Northeast",
                       ifelse(state$State %in% midwest_states, "Midwest",
                              ifelse(state$State %in% south_states, "South",
                                     ifelse(state$State %in% west_states, "West", NA))))

Visualizations

#Loading package
library(ggplot2)

#Exploring trends over time plot
ggplot(cleaned_data, aes(x = Year, y = Food.secure.percent)) +
  geom_line(color = "blue") +
  geom_line(aes(y = Food.insecure.percent), color = "red") +
  labs(title = "Trends in Food Security Over Time",
       x = "Year",
       y = "Percentage",
       color = "Status") +
  scale_color_manual(values = c("blue", "red"),
                     labels = c("Food Secure", "Food Insecure")) +
  theme_minimal()

#Plotting food insecurity by household composition
ggplot(cleaned_data, aes(x = Subcategory, y = Food.insecure.percent)) +
  geom_bar(stat = "identity", fill = "skyblue") +
  labs(title = "Food Insecurity by Household Composition",
       x = "Household Composition",
       y = "Percentage of Food Insecure",
       fill = "Status") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

#Analyzing overall food security status
summary(cleaned_data[, c("Food.secure.percent", "Food.insecure.percent", "Low.food.security.percent", "Very.low.food.security.percent")])

##  Food.secure.percent Food.insecure.percent Low.food.security.percent
##  Min.   :57.00       Min.   : 4.90         Min.   : 3.200           
##  1st Qu.:81.50       1st Qu.: 9.80         1st Qu.: 5.900           
##  Median :87.30       Median :12.70         Median : 7.800           
##  Mean   :84.26       Mean   :15.74         Mean   : 9.984           
##  3rd Qu.:90.20       3rd Qu.:18.50         3rd Qu.:13.200           
##  Max.   :95.10       Max.   :43.00         Max.   :25.300           
##  Very.low.food.security.percent
##  Min.   : 0.800                
##  1st Qu.: 3.600                
##  Median : 4.700                
##  Mean   : 5.752                
##  3rd Qu.: 6.500                
##  Max.   :19.300

Plots showing the trends in food security over time and another displaying food insecurity by household composition.

#Creating a scatter plot
ggplot(state, aes(x = Food.insecurity.prevalence, y = Very.low.food.security.prevalence)) +
  geom_point() +
  labs(title = "Food Insecurity vs. Very Low Food Security",
       x = "Food Insecurity Prevalence (%)",
       y = "Very Low Food Security Prevalence (%)") +
  theme_minimal()

In this scatter plot we visualize the relationship between food insecurity and very low food security prevalence across states.

#Checking the dplyr package 
if (!requireNamespace("dplyr", quietly = TRUE)) {
  install.packages("dplyr")
}
library(dplyr)

## 
## 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

#Calculating average food insecurity prevalence by region
average_food_insecurity <- state %>%
  group_by(Region) %>%
  summarize(Average_Food_Insecurity_Prevalence = mean(Food.insecurity.prevalence, na.rm = TRUE))

#Visualization the disparities
ggplot(average_food_insecurity, aes(x = Region, y = Average_Food_Insecurity_Prevalence)) +
  geom_bar(stat = "identity", fill = "skyblue") +
  labs(title = "Average Food Insecurity Prevalence by Region",
       x = "Region",
       y = "Average Food Insecurity Prevalence (%)") +
  theme_minimal()

#Extracting starting year from the Year column
state$Start_Year <- as.numeric(substring(state$Year, 1, 4))

#Calculating average very low food security prevalence by region and starting year
average_very_low_food_security_by_start_year <- state %>%
  group_by(Region, Start_Year) %>%
  summarize(Average_Very_Low_Food_Security_Prevalence = mean(Very.low.food.security.prevalence, na.rm = TRUE))

## `summarise()` has grouped output by 'Region'. You can override using the
## `.groups` argument.

#Visualization the trend by region
ggplot(average_very_low_food_security_by_start_year, aes(x = Start_Year, y = Average_Very_Low_Food_Security_Prevalence, color = Region)) +
  geom_line() +
  labs(title = "Trend of Very Low Food Security Prevalence by Region",
       x = "Year",
       y = "Average Very Low Food Security Prevalence (%)",
       color = "Region") +
  theme_minimal() +
  theme(legend.position = "top")

The decrease in very low food security prevalence from 2015 to 2020 may be attributed to overall improvements in economic conditions during that period. Factors such as job growth, wage increases, and government assistance programs could have contributed to reduced food insecurity levels.

References

Economic Research Service U.S. DEPARTMENT OF AGRICULTURE https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-u-s/interactive-charts-and-highlights/