Data Source: http://www.childmortality.org/
Data Set:“Estimates for under-five, infant and neonatal mortality”
The dataset contains six value variables of interest: under-five (0-4 years) mortality, infant (0-1 years) mortality, neonatal (0-1 month) mortality, as well as the number of under-five, infant, and neonatal deaths.
# install.packages("tidyr")
# install.packages("dplyr")
# install.packages("ggplot2")
library(tidyr)
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, unionlibrary(ggplot2)
library(stringr)death_rate<- read.csv("https://raw.githubusercontent.com/olgashiligin/project2_607/master/Child_Death_Rate.csv")changing wide structure to long
splitting year_type column into year and type columns
filtering data for the last 20 years with Uncertainty bound = “Median”
selecting necessary columns
median_bound<-death_rate%>%
gather(year_type, value, U5MR.1950:Neonatal.Deaths.2015) %>%
mutate(year = sub('.*\\.', '',year_type), type = sub('\\.\\d{4}', '', year_type)) %>%
filter(Uncertainty.bounds. == "Median"& year>=1998) %>%
select(-c(year_type,Uncertainty.bounds.))
head(median_bound)## ISO.Code CountryName value year type
## 1 AFG Afghanistan 142.6 1998 U5MR
## 2 AGO Angola 220.8 1998 U5MR
## 3 ALB Albania 28.9 1998 U5MR
## 4 AND Andorra 5.0 1998 U5MR
## 5 ARE United Arab Emirates 11.8 1998 U5MR
## 6 ARG Argentina 21.4 1998 U5MRstats<-median_bound %>%
group_by(type) %>%
summarise(avg_value = mean(value))
head(stats)## # A tibble: 6 x 2
## type avg_value
## <chr> <dbl>
## 1 IMR 32.1
## 2 Infant.Deaths 29360.
## 3 Neonatal.Deaths 17090.
## 4 NMR 17.2
## 5 U5MR 45.2
## 6 Under.five.Deaths 40965.For the last 20 years Death Under Five group has the highest average MR - 45, whereas the lowest MR has group Neonatal.Deaths - 17.
What country has the highest and the lowest under-five MR?
U5MR<-median_bound %>%
filter(type=="U5MR") %>%
arrange(desc(value))
head(U5MR)## ISO.Code CountryName value year type
## 1 SLE Sierra Leone 246.0 1998 U5MR
## 2 NER Niger 245.8 1998 U5MR
## 3 SLE Sierra Leone 241.1 1999 U5MR
## 4 NER Niger 236.4 1999 U5MR
## 5 SLE Sierra Leone 235.8 2000 U5MR
## 6 RWA Rwanda 234.0 1998 U5MRtail(U5MR)## ISO.Code CountryName value year type
## 3505 ISL Iceland 2.1 2013 U5MR
## 3506 ISL Iceland 2.1 2014 U5MR
## 3507 LUX Luxembourg 2.0 2013 U5MR
## 3508 LUX Luxembourg 2.0 2014 U5MR
## 3509 ISL Iceland 2.0 2015 U5MR
## 3510 LUX Luxembourg 1.9 2015 U5MRWhat countries have the highest and the lowest IMR?
IMR<-median_bound %>%
filter(type=="IMR") %>%
arrange(desc(value))
head(IMR)## ISO.Code CountryName value year type
## 1 SLE Sierra Leone 148.1 1998 IMR
## 2 SLE Sierra Leone 145.8 1999 IMR
## 3 SLE Sierra Leone 143.3 2000 IMR
## 4 SLE Sierra Leone 140.5 2001 IMR
## 5 LBR Liberia 138.1 1998 IMR
## 6 SLE Sierra Leone 137.7 2002 IMRtail(IMR)## ISO.Code CountryName value year type
## 3505 ISL Iceland 1.6 2013 IMR
## 3506 LUX Luxembourg 1.6 2013 IMR
## 3507 ISL Iceland 1.6 2014 IMR
## 3508 LUX Luxembourg 1.6 2014 IMR
## 3509 ISL Iceland 1.6 2015 IMR
## 3510 LUX Luxembourg 1.5 2015 IMRWhat countries have the highest and the lowest NMR?
NMR<-median_bound %>%
filter(type=="NMR") %>%
arrange(desc(value))
head(NMR)## ISO.Code CountryName value year type
## 1 PAK Pakistan 63.2 1998 NMR
## 2 PAK Pakistan 61.8 1999 NMR
## 3 MLI Mali 61.1 1998 NMR
## 4 PAK Pakistan 60.4 2000 NMR
## 5 SSD South Sudan 60.1 1998 NMR
## 6 MLI Mali 59.1 1999 NMRtail(NMR)## ISO.Code CountryName value year type
## 3505 SMR San Marino 0.8 2010 NMR
## 3506 SMR San Marino 0.8 2011 NMR
## 3507 SMR San Marino 0.8 2012 NMR
## 3508 SMR San Marino 0.7 2013 NMR
## 3509 SMR San Marino 0.7 2014 NMR
## 3510 SMR San Marino 0.7 2015 NMRWhat year have 3 types of child death had the highest number in?
stats_by_year<-median_bound %>%
group_by(type, year) %>%
summarise(avg_value = mean(value)) %>%
group_by(type, year) %>%
summarise(max_avg_value = max(avg_value)) %>%
arrange(year,desc(max_avg_value)) %>%
slice(1)
stats_by_year## # A tibble: 6 x 3
## # Groups: type [6]
## type year max_avg_value
## <chr> <chr> <dbl>
## 1 IMR 1998 41.9
## 2 Infant.Deaths 1998 36468.
## 3 Neonatal.Deaths 1998 20715.
## 4 NMR 1998 21.4
## 5 U5MR 1998 61.2
## 6 Under.five.Deaths 1998 52641.As we see 1988 was the worst year in terms of child death among all 3 groups of child death for the last 20 years.
Let’s pick 3 countrise, one from developed countries (United Kingdom), one from economies in transition (Belarus) and one from undeveloped (Angola) counrties in order to compare child death rates (Classifucation was taken from United Nations Country Classification).
United Kingdom
GB<-median_bound %>%
filter(CountryName=="United Kingdom") %>%
group_by(type) %>%
summarise(avg_value = mean(value))
GB## # A tibble: 6 x 2
## type avg_value
## <chr> <dbl>
## 1 IMR 4.8
## 2 Infant.Deaths 3617.
## 3 Neonatal.Deaths 2457.
## 4 NMR 3.26
## 5 U5MR 5.67
## 6 Under.five.Deaths 4264.Belarus
Belarus<-median_bound %>%
filter(CountryName=="Belarus") %>%
group_by(type) %>%
summarise(avg_value = mean(value))
Belarus## # A tibble: 6 x 2
## type avg_value
## <chr> <dbl>
## 1 IMR 7.14
## 2 Infant.Deaths 715.
## 3 Neonatal.Deaths 419.
## 4 NMR 4.22
## 5 U5MR 9.12
## 6 Under.five.Deaths 907Angola
Angola<-median_bound %>%
filter(CountryName=="Angola") %>%
group_by(type) %>%
summarise(avg_value = mean(value))
Angola## # A tibble: 6 x 2
## type avg_value
## <chr> <dbl>
## 1 IMR 116.
## 2 Infant.Deaths 102518.
## 3 Neonatal.Deaths 48698.
## 4 NMR 54.7
## 5 U5MR 194.
## 6 Under.five.Deaths 169552Results are quite expected and show negative correlation between child mortality variables and country’s economic development. (Ignore absolute death numbers).