DATA 607 Project 2 Part 2

UNICEF dataset on Under 5 Mortality

Discussion thread created by : Samuel Bellows

1. Introduction

UNICEF dataset that gives the under 5 mortality for many countries across the years 1950-2015. The problem is that the year variable is spread out into 65 different columns, 1 for each year, that need to be gathered into 1 column. In order to make this dataset tiny we would gather the year columns into one column until we had a 3 column dataset of Country name, Year, and Mortality.

2. Load library

#install.packages("dplyr")
#install.packages("tidyr")
#install.packages("ggplot2")
#install.packages("DT")
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

library(tidyr)
library(ggplot2)
library(DT)

3. Data load and cleaning

Data is stored in the Github and loaded data from Github to Rstudio using read.csv() method.

# read csv file data in a variable using  read.csv() 
data <- read.csv('https://raw.githubusercontent.com/SubhalaxmiRout002/Data-607-Project-2-Dataset-1/master/unicef-u5mr.csv', header = TRUE, stringsAsFactors = F)

# convert data to data frame
data <- data.frame(data)

# display data using datatable
datatable(data,options = list(scrollX = TRUE, paging=TRUE,fixedHeader=TRUE))

3.1 Gather year from 1950 to 2015

This dataset year has given from 1950 to 2015. Each year mentioned as a column. Using tidyr convert these columns to Year column.

# using gather() convert column to row
data <- data %>% gather(Year, Mortality, U5MR.1950:U5MR.2015, na.rm = TRUE)

# Uremove "U5MR" from the name
data$Year  <- sub('U5MR.','',data$Year)

# arrange Mortality by desc order
data <- data %>% arrange(desc(Mortality))

# display data using datatable
datatable(data,options = list(scrollX = TRUE, paging=TRUE,fixedHeader=TRUE))

4. Analysis

Mortality based on year:

• Year: 1950, Highest/lowest mortality
• Year: 2015 Highest/lowest mortality

4.1 Highest mortality in the year 1950

# apply filter on year = 1950 
data1 <- data %>% filter(Year == "1950")

# display data using datatable
datatable(data1,options = list(scrollX = TRUE, paging=TRUE,fixedHeader=TRUE))

4.2 Highest to lowest mortality in 1950

data1 <- data.frame(data1)

ggplot(data = data1, mapping = aes(x = CountryName, y = Mortality)) + 
  geom_bar(aes(reorder(CountryName,Mortality),Mortality),stat = "identity",fill = "steelblue") +
  coord_flip() + xlab("Country") + ylab("#Mortality") +
  ggtitle("Highest to lowest mortality in 1950") +
  theme(plot.title = element_text(hjust = 0.5),panel.background = element_rect(fill = "white", color = NA)) +
  geom_text(aes( y = Mortality,label=Mortality), hjust = -0.20, color="black", size=3.5)

4.3 Highest mortality in 2015

# apply filter on year = 2015 
data2 <- data %>% filter(Year == "2015")

# arrange Mortality by desc
data2 <- data2 %>% arrange(desc(Mortality))

# display data using datatable
datatable(data2,options = list(scrollX = TRUE, paging=TRUE,fixedHeader=TRUE))

4.4 Top 10 countries with mortality in 2015

data3 <- data.frame(head(data2, 10))

ggplot(data = data3, mapping = aes(x = CountryName, y = Mortality)) + 
  geom_bar(aes(reorder(CountryName,Mortality),Mortality),stat = "identity",fill = "steelblue") +
  coord_flip() + xlab("Country") + ylab("#Mortality") +
  ggtitle("Top 10 countries with mortality in 2015") +
  theme(plot.title = element_text(hjust = 0.5),panel.background = element_rect(fill = "white", color = NA)) +
  geom_text(aes( y = Mortality,label=Mortality), hjust=-0.20, color="black", size=3.5)

4.5 Bottom 10 countries with mortality in 2015

# last 10 row data stored in the variable
data4 <- data.frame(tail(data2, 10))

ggplot(data = data4, mapping = aes(x = CountryName, y = Mortality)) + 
  geom_bar(aes(reorder(CountryName,Mortality),Mortality),stat = "identity",fill = "steelblue") +
  coord_flip() + xlab("Country") + ylab("#Mortality") +
  ggtitle("Bottom 10 countries with mortality  in 2015") +
  theme(plot.title = element_text(hjust = 0.5),panel.background = element_rect(fill = "white", color = NA)) +
  geom_text(aes( y = Mortality,label=Mortality), hjust=-0.20,color="black", size=3.5)

4.6 Average mortality of the 10 developed countries

# stored data in data frame
data5 <- data.frame(data)

# calculate mean and group by using country
data5 <- data5 %>%  group_by(CountryName) %>%  summarise(round(mean(Mortality),2))
data5 <- mutate(data5,Avg_mortality = `round(mean(Mortality), 2)`)

# select country name and average mortality
data5 <- select(data5, CountryName, Avg_mortality)

#apply filter on country
data5 <- data5 %>% filter(grepl('United States of America|Japan|Norway|Switzerland|Australia|Ireland|Germany|Iceland|Singapore|Sweden|Netherlands' , CountryName  )) %>% arrange(desc(Avg_mortality))


# 10 developed countries average mortality
ggplot(data = data5, mapping = aes(x = CountryName, y = Avg_mortality)) + 
  geom_bar(aes(reorder(CountryName,Avg_mortality),Avg_mortality),stat = "identity",fill = "steelblue") +
   xlab("Country") + ylab("Average Mortality") +
  ggtitle("Developed countries  Average Mortality") +
  theme(plot.title = element_text(hjust = 0.5)) +
  geom_text(aes( y = Avg_mortality,label=Avg_mortality), vjust = - 0.2, color="black", size=3.5) +
  theme(axis.text.x=element_text(angle=30,hjust=1),panel.background = element_rect(fill = "white", color = NA))

4.7 Mortality of United States from 1950 to 2015

# store data in data frame
data6 <- data.frame(data)

# apply filter on Contry name = US
data6 <- data6 %>% filter(CountryName == "United States of America")


ggplot(data6,aes(x = CountryName, y = Mortality)) + 
  geom_boxplot() +  xlab("Country") +

stat_summary(fun=mean, colour="darkred", geom="point", size=3,show.legend = FALSE) +
stat_summary(fun=mean, colour="red", geom="text", show.legend = FALSE, 
               vjust=-0.7, aes( label=round(..y.., digits=1))) 

# display mean, median, upper hinge, lower hindge
summary(data6$Mortality)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   6.500   8.625  13.650  17.467  26.225  37.700

5. Conclusion

The plot 4.2, plot 4.4, and plot 4.5 shows mortality goes down year over year.

• In the year 1950, the African continent countries have more mortality.
• In the year 2015, developed countries have fewer mortality as compared to undeveloped or developing countries.
• The average mortality for 10 developed countries: Sweden has the least average and Singapore has the highest average.
• The boxplot (plot 4.7) shows the mortality of the United States from 1950 to 2015.
• The mortality of the United States has gone down year over year.