Malaria Deaths

This data exploratory project is part of TutorialTuesday

(https://www.linkedin.com/pulse/november-data-challenge-kate-strachnyi-/)

and TidyTuesday

(https://github.com/rfordatascience/tidytuesday)

Load libraries

library(tidyverse)
library(lubridate)
library(skimr)
library(RColorBrewer)
library(scales)

Import datasets

malaria_deaths = read.csv("malaria_deaths_by_age.csv")
# https://www.gapminder.org/data/geo/
geography = read.csv("geography.csv")

Explore our data

malaria_deaths %>% glimpse()

Observations: 30,780
Variables: 6
$ X         <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30...
$ entity    <fct> Afghanistan, Afghanistan, Afghanistan, Afghanistan, Afghanistan, Afghanistan, Afghanistan, Afghanistan, Afgha...
$ code      <fct> AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG, AFG,...
$ year      <int> 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2...
$ age_group <fct> Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, U...
$ deaths    <dbl> 184.6064, 191.6582, 197.1402, 207.3578, 226.2094, 236.3280, 250.8689, 257.6679, 273.0083, 272.2882, 278.3776,...

geography %>% glimpse()

Observations: 197
Variables: 11
$ geo                          <fct> afg, alb, dza, and, ago, atg, arg, arm, aus, aut, aze, bhs, bhr, bgd, brb, blr, bel, blz, ...
$ name                         <fct> Afghanistan, Albania, Algeria, Andorra, Angola, Antigua and Barbuda, Argentina, Armenia, A...
$ four_regions                 <fct> asia, europe, africa, europe, africa, americas, americas, europe, asia, europe, europe, am...
$ eight_regions                <fct> asia_west, europe_east, africa_north, europe_west, africa_sub_saharan, america_north, amer...
$ six_regions                  <fct> south_asia, europe_central_asia, middle_east_north_africa, europe_central_asia, sub_sahara...
$ members_oecd_g77             <fct> g77, others, g77, others, g77, g77, g77, others, oecd, oecd, others, g77, g77, g77, g77, o...
$ Latitude                     <dbl> 33.00000, 41.00000, 28.00000, 42.50779, -12.50000, 17.05000, -34.00000, 40.25000, -25.0000...
$ Longitude                    <dbl> 66.00000, 20.00000, 3.00000, 1.52109, 18.50000, -61.80000, -64.00000, 45.00000, 135.00000,...
$ UN.member.since              <fct> 19/11/1946, 14/12/1955, 8/10/62, 28/7/1993, 1/12/76, 11/11/81, 24/10/1945, 2/3/92, 1/11/45...
$ World.bank.region            <fct> South Asia, Europe & Central Asia, Middle East & North Africa, Europe & Central Asia, Sub-...
$ World.bank.income.group.2017 <fct> Low income, Upper middle income, Upper middle income, High income, Lower middle income, Hi...

Let’s clean up the data and rename some of the columns

malaria_deaths$X = NULL
colnames(malaria_deaths) = c("Country", "Code", "Year", "Age_Group", "Deaths")

Review our datasets before performing joining

summary(malaria_deaths)

                 Country           Code            Year            Age_Group        Deaths        
 Afghanistan         :  135   AFG    :  135   Min.   :1990   15-49      :6156   Min.   :     0.0  
 Albania             :  135   AGO    :  135   1st Qu.:1996   5-14       :6156   1st Qu.:     0.0  
 Algeria             :  135   ALB    :  135   Median :2003   50-69      :6156   Median :     0.1  
 American Samoa      :  135   AND    :  135   Mean   :2003   70 or older:6156   Mean   :  3698.6  
 Andean Latin America:  135   ARE    :  135   3rd Qu.:2010   Under 5    :6156   3rd Qu.:    80.5  
 Andorra             :  135   (Other):25785   Max.   :2016                      Max.   :752025.5  
 (Other)             :29970   NA's   : 4320                                                       

Return the number of countries in the “maria_deaths”

malaria_deaths %>% 
  select(Country) %>%
  unique() %>% 
  count()

Make the “Code” column lowercase. We do this as eventually we want to merge this one with the “geography” dataset

malaria_deaths$Code = tolower(malaria_deaths$Code)

Return the number of countries in “geography”

geography %>% select(name) %>% count() 

More data manipulation for “geography”

colnames(geography)[1:2] <- c("Code", "Country")
glimpse(geography)

Observations: 197
Variables: 11
$ Code                         <fct> afg, alb, dza, and, ago, atg, arg, arm, aus, aut, aze, bhs, bhr, bgd, brb, blr, bel, blz, ...
$ Country                      <fct> Afghanistan, Albania, Algeria, Andorra, Angola, Antigua and Barbuda, Argentina, Armenia, A...
$ four_regions                 <fct> asia, europe, africa, europe, africa, americas, americas, europe, asia, europe, europe, am...
$ eight_regions                <fct> asia_west, europe_east, africa_north, europe_west, africa_sub_saharan, america_north, amer...
$ six_regions                  <fct> south_asia, europe_central_asia, middle_east_north_africa, europe_central_asia, sub_sahara...
$ members_oecd_g77             <fct> g77, others, g77, others, g77, g77, g77, others, oecd, oecd, others, g77, g77, g77, g77, o...
$ Latitude                     <dbl> 33.00000, 41.00000, 28.00000, 42.50779, -12.50000, 17.05000, -34.00000, 40.25000, -25.0000...
$ Longitude                    <dbl> 66.00000, 20.00000, 3.00000, 1.52109, 18.50000, -61.80000, -64.00000, 45.00000, 135.00000,...
$ UN.member.since              <fct> 19/11/1946, 14/12/1955, 8/10/62, 28/7/1993, 1/12/76, 11/11/81, 24/10/1945, 2/3/92, 1/11/45...
$ World.bank.region            <fct> South Asia, Europe & Central Asia, Middle East & North Africa, Europe & Central Asia, Sub-...
$ World.bank.income.group.2017 <fct> Low income, Upper middle income, Upper middle income, High income, Lower middle income, Hi...

Inner join “malaria_deaths” onto geography

malaria_deaths = malaria_deaths %>%
  inner_join(geography, by = "Code")

Column `Code` joining character vector and factor, coercing into character vector

colnames(malaria_deaths)

 [1] "Country.x"                    "Code"                         "Year"                         "Age_Group"                   
 [5] "Deaths"                       "Country.y"                    "four_regions"                 "eight_regions"               
 [9] "six_regions"                  "members_oecd_g77"             "Latitude"                     "Longitude"                   
[13] "UN.member.since"              "World.bank.region"            "World.bank.income.group.2017"

Return some descriptive stats for the joined dataframe

summary(malaria_deaths)

               Country.x         Code                Year            Age_Group        Deaths                        Country.y    
 Afghanistan        :  135   Length:25380       Min.   :1990   15-49      :5076   Min.   :     0.00   Afghanistan        :  135  
 Albania            :  135   Class :character   1st Qu.:1996   5-14       :5076   1st Qu.:     0.00   Albania            :  135  
 Algeria            :  135   Mode  :character   Median :2003   50-69      :5076   Median :     0.07   Algeria            :  135  
 Andorra            :  135                      Mean   :2003   70 or older:5076   Mean   :   917.75   Andorra            :  135  
 Angola             :  135                      3rd Qu.:2010   Under 5    :5076   3rd Qu.:    44.40   Angola             :  135  
 Antigua and Barbuda:  135                      Max.   :2016                      Max.   :261794.56   Antigua and Barbuda:  135  
 (Other)            :24570                                                                            (Other)            :24570  
   four_regions             eight_regions                    six_regions   members_oecd_g77    Latitude        Longitude      
 africa  :7290   africa_sub_saharan:6480   america                 :4590         :  135     Min.   :-42.00   Min.   :-175.00  
 americas:4590   east_asia_pacific :3780   east_asia_pacific       :3780   g77   :17415     1st Qu.:  4.00   1st Qu.:  -6.50  
 asia    :7425   asia_west         :3645   europe_central_asia     :6750   oecd  : 4050     Median : 17.27   Median :  21.88  
 europe  :6075   europe_east       :3240   middle_east_north_africa:2700   others: 3780     Mean   : 18.92   Mean   :  21.00  
                 america_north     :2970   south_asia              :1080                    3rd Qu.: 39.75   3rd Qu.:  48.64  
                 europe_west       :2835   sub_saharan_africa      :6480                    Max.   : 65.00   Max.   : 178.00  
                 (Other)           :2430                                                                                      
   UN.member.since                   World.bank.region      World.bank.income.group.2017
 24/10/1945: 3645   Europe & Central Asia     :6615                       :   0         
 14/12/1955: 2160   Sub-Saharan Africa        :6480    High income        :7020         
 20/9/1960 : 1890   Latin America & Caribbean :4320    Low income         :4185         
 2/3/92    : 1080   East Asia & Pacific       :3780    Lower middle income:7020         
 17/9/1991 :  945   Middle East & North Africa:2835    Upper middle income:7155         
 18/9/1962 :  540   South Asia                :1080                                     
 (Other)   :15120   (Other)                   : 270                                     

More data manipulation

colnames(malaria_deaths)[1] = "Country"
colnames(malaria_deaths)[15] = "Income_Group"
colnames(malaria_deaths)[7] = "Four_Reg"
colnames(malaria_deaths)[9] = "Six_Reg"

Subset the joined dataset to get only the variables we want

malaria_deaths = 
  malaria_deaths[c("Country", "Year", "Age_Group", "Deaths", "Four_Reg", "Six_Reg",
                   "Latitude", "Longitude", "Income_Group", "Latitude", "Longitude")]

summary(malaria_deaths)

                Country           Year            Age_Group        Deaths              Four_Reg                        Six_Reg    
 Afghanistan        :  135   Min.   :1990   15-49      :5076   Min.   :     0.00   africa  :7290   america                 :4590  
 Albania            :  135   1st Qu.:1996   5-14       :5076   1st Qu.:     0.00   americas:4590   east_asia_pacific       :3780  
 Algeria            :  135   Median :2003   50-69      :5076   Median :     0.07   asia    :7425   europe_central_asia     :6750  
 Andorra            :  135   Mean   :2003   70 or older:5076   Mean   :   917.75   europe  :6075   middle_east_north_africa:2700  
 Angola             :  135   3rd Qu.:2010   Under 5    :5076   3rd Qu.:    44.40                   south_asia              :1080  
 Antigua and Barbuda:  135   Max.   :2016                      Max.   :261794.56                   sub_saharan_africa      :6480  
 (Other)            :24570                                                                                                        
    Latitude        Longitude                    Income_Group    Latitude.1      Longitude.1     
 Min.   :-42.00   Min.   :-175.00                      :   0   Min.   :-42.00   Min.   :-175.00  
 1st Qu.:  4.00   1st Qu.:  -6.50   High income        :7020   1st Qu.:  4.00   1st Qu.:  -6.50  
 Median : 17.27   Median :  21.88   Low income         :4185   Median : 17.27   Median :  21.88  
 Mean   : 18.92   Mean   :  21.00   Lower middle income:7020   Mean   : 18.92   Mean   :  21.00  
 3rd Qu.: 39.75   3rd Qu.:  48.64   Upper middle income:7155   3rd Qu.: 39.75   3rd Qu.:  48.64  
 Max.   : 65.00   Max.   : 178.00                              Max.   : 65.00   Max.   : 178.00  
                                                                                                 

Plot a bar plot to show total number of deaths by age group for each region

ggplot(malaria_deaths) +
  geom_bar(stat = "identity", width = 1.5, aes(x = Four_Reg, y = Deaths, fill = Age_Group), position = "dodge") +
  scale_y_continuous(limits = c(0, 250000)) +
  ggtitle("Total Number of Malaria Deaths for Different Age Groups by Region (1990 - 2016)") +
  theme_minimal() +
  xlab("World Regions") +
  ylab("Number of Deaths") +
  theme(axis.text.y = element_text(size = 12, face = 'bold'),
        axis.text.x = element_text(size = 12, face = 'bold'),
        axis.title.y = element_text(size = 12, face = 'bold'),
        axis.title.x = element_text(size = 12, face = 'bold'),
        plot.title = element_text(size = 12, face = "bold")) +
  scale_fill_brewer(palette = "Set1")

It seems Africa exhibits an extremely high number of deaths. To see the picture more clearly, let’s look at Africa and the rest of the world separately

africa = 
  malaria_deaths %>%
  filter(Four_Reg == "africa")

ggplot(africa) +
  geom_bar(stat = "identity", aes(x = Four_Reg, y = Deaths, fill = Age_Group), position = "dodge") +
  scale_y_continuous(limits = c(0, 250000)) +
  ggtitle("Total Number of Malaria Deaths for Different Age Groups in Africa (1990 - 2016)") +
  theme_minimal() +
  xlab("Africa") +
  ylab("Number of Deaths") +
  theme(axis.text.y = element_text(size = 12, face = 'bold'),
        axis.text.x = element_text(size = 12, face = 'bold'),
        axis.title.y = element_text(size = 12, face = 'bold'),
        axis.title.x = element_text(size = 12, face = 'bold'),
        plot.title = element_text(size = 12, face = "bold"),
        panel.grid.major = element_blank()) +
  scale_fill_brewer(palette = "Set1")

We can see that the group affected the most in Africa is the “Under 5” group. Now, let’s look at the rest of the world.

world_no_africa = 
  malaria_deaths %>%
  filter(Four_Reg != "africa")

ggplot(world_no_africa) +
  geom_bar(stat = "identity", width = 0.8, aes(x = Four_Reg, y = Deaths, fill = Age_Group), position = "dodge") +
  scale_y_continuous(limits = c(0, 50000)) +
  ggtitle("Total Number of Malaria Deaths for Different Age Groups by Region (1990 - 2016)") +
  theme_minimal() +
  xlab("World Regions") +
  ylab("Number of Deaths") +
  theme(axis.text.y = element_text(size = 12, face = 'bold'),
        axis.text.x = element_text(size = 12, face = 'bold'),
        axis.title.y = element_text(size = 12, face = 'bold'),
        axis.title.x = element_text(size = 12, face = 'bold'),
        plot.title = element_text(size = 12, face = "bold")) +
  scale_fill_brewer(palette = "Set1")

We can see that the death toll in Americas and Europe is close to zero. The group affected the most in Asia is also the “Under 5” group.

Now, let’s look at the trend of the country with the most number of malaria deaths

malaria_deaths %>%
  arrange(desc(Deaths))

The country with the highest number of malaria deaths over the year is Nigeria. Let’s zoom into Nigeria and see the trend over the year

nigeria = 
  malaria_deaths %>%
  filter(Country == "Nigeria")
head(nigeria)

Explore our new filtered dataset

glimpse(nigeria)

Observations: 135
Variables: 11
$ Country      <fct> Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeria, Nigeri...
$ Year         <int> 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 200...
$ Age_Group    <fct> Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under 5, Under ...
$ Deaths       <dbl> 169612.234, 177084.049, 182532.594, 189569.206, 194656.088, 197338.651, 201157.357, 203241.332, 209637.81...
$ Four_Reg     <fct> africa, africa, africa, africa, africa, africa, africa, africa, africa, africa, africa, africa, africa, a...
$ Six_Reg      <fct> sub_saharan_africa, sub_saharan_africa, sub_saharan_africa, sub_saharan_africa, sub_saharan_africa, sub_s...
$ Latitude     <dbl> 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 1...
$ Longitude    <dbl> 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, ...
$ Income_Group <fct> Lower middle income, Lower middle income, Lower middle income, Lower middle income, Lower middle income, ...
$ Latitude.1   <dbl> 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 1...
$ Longitude.1  <dbl> 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, ...

summary(nigeria)

                 Country         Year            Age_Group      Deaths           Four_Reg                       Six_Reg   
 Nigeria             :135   Min.   :1990   15-49      :27   Min.   :  3141   africa  :135   america                 :  0  
 Afghanistan         :  0   1st Qu.:1996   5-14       :27   1st Qu.:  5808   americas:  0   east_asia_pacific       :  0  
 Albania             :  0   Median :2003   50-69      :27   Median :  9238   asia    :  0   europe_central_asia     :  0  
 Algeria             :  0   Mean   :2003   70 or older:27   Mean   : 50235   europe  :  0   middle_east_north_africa:  0  
 American Samoa      :  0   3rd Qu.:2010   Under 5    :27   3rd Qu.: 20377                  south_asia              :  0  
 Andean Latin America:  0   Max.   :2016                    Max.   :261795                  sub_saharan_africa      :135  
 (Other)             :  0                                                                                                 
    Latitude    Longitude              Income_Group   Latitude.1  Longitude.1
 Min.   :10   Min.   :8                      :  0   Min.   :10   Min.   :8   
 1st Qu.:10   1st Qu.:8   High income        :  0   1st Qu.:10   1st Qu.:8   
 Median :10   Median :8   Low income         :  0   Median :10   Median :8   
 Mean   :10   Mean   :8   Lower middle income:135   Mean   :10   Mean   :8   
 3rd Qu.:10   3rd Qu.:8   Upper middle income:  0   3rd Qu.:10   3rd Qu.:8   
 Max.   :10   Max.   :8                             Max.   :10   Max.   :8   
                                                                             

Plot the number of malaria deaths in Nigeria over the period 1990 - 2016 for different age groups

ggplot(nigeria, aes(x = Year, y = Deaths)) +
  geom_line(aes(color = Age_Group)) +
  ggtitle("Total Number of Malaria Deaths for Different Age Groups in Nigeria (1990 - 2016)") +
  theme_minimal() +
  xlab("Year") +
  ylab("Number of Deaths") +
  theme(axis.text.y = element_text(size = 12, face = 'bold'),
        axis.text.x = element_text(size = 12, face = 'bold'),
        axis.title.y = element_text(size = 12, face = 'bold'),
        axis.title.x = element_text(size = 12, face = 'bold'),
        plot.title = element_text(size = 12, face = "bold", hjust = 0.2),
        panel.grid.major = element_blank()) +
  scale_color_brewer(palette = "Set1") +
  scale_x_continuous(breaks = pretty_breaks(n = 10)) +
  scale_y_continuous(breaks = pretty_breaks(n = 5))