$1.$ Make all the obvious visual interpretation and explain each boxplot in detail.

$2.$ Do statistical tests between groups that are not very obviously different:

$-$ i$.$e$.2$ sample t test between europe and africa, anova for all the groups

$-$ Remember to log transform or remove outliers to make tests more accurate

CODE IN R:$-$

library$($tidyverse$)$

library$($ggplot$2)$

# Change the file directories!!

country$\_$metadata $<-$ read$\_$csv$(\text{'}$country$\_$metadata.csv$\text{'})$

education $<-$ education $\%>\%$

pivot$\_$longer$($cols $=5$:$68,$ names$\_$to $=$ 'year'$)\%>\%$

full$\_$join$($country$\_$metadata$[1$:$3])\%>\%$

filter$(!$is$.$na$($value$))\%>\%$

filter$(!$is$.$na$($Region$))\%>\%$

filter$(!$is$.$na$($IncomeGroup$))$ # Run this again to filter venezuela

# Log transform to handle outliers

education$log$\_$value $<-$ log$($education$value$)$

education$\_$decade $<-$ education $\%>\%$

filter$($year $\%$in$\%$ c$(2013$:$2022))$

# Label data as male or female, filter data pertaining to gender

education$\_$mf $<-$ education $\%>\%$

mutate$($male$\_$or$\_$female $=$ case$\_$when$($

str$\_$detect$(`$Indicator Name$`,\text{'}$ male'$)$ ~ 'male',

str$\_$detect$(`$Indicator Name$`,$ 'female'$)$ ~ 'female'

$)$

$)\%>\%$

filter$(!$is$.$na$($male$\_$or$\_$female$))$

# Perform $2-$sample t$-$test between Europe and Africa

t$\_$test$\_$result $<-$ t$.$test$($education$\_$mf$value ~ education$\_$mf$Region,

subset$($education$\_$mf$,$ Region $\%$in$\%$ c$("$Europe$",$ "Africa"$)))$

# Perform ANOVA for all groups

anova$\_$result $<-$ aov$($value ~ Region, data $=$ education$\_$mf$)$

# Print results

print$("2-$Sample t$-$test between Europe and Africa:"$)$

print$($t$\_$test$\_$result$)$

print$("$ANOVA Test for all groups:"$)$

print$($anova$\_$result$)$

# Indicator: Adjusted net enrollment by income group and region

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Adjusted net enrollment rate'$))\%>\%$

ggplot$()+$ geom$\_$boxplot$($aes$($x $=$ value, fill $=$ IncomeGroup$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Adjusted net enrollment rate by income group'$)+$

xlab$(\text{'}$Adjusted net enrollment $\%\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Adjusted net enrollment rate'$))\%>\%$

ggplot$()+$ geom$\_$boxplot$($aes$($x $=$ value, fill $=$ Region$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Adjusted net enrollment rate by region'$)+$

xlab$(\text{'}$Adjusted net enrollment $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

# Indicator: Repeaters by income group and region

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Repeaters'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ IncomeGroup$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Repeaters by income group'$)+$

xlab$(\text{'}$Repeaters $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Repeaters'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ Region$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Repeaters by region'$)+$

xlab$(\text{'}$Repeaters $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

# Indicator: Trained teachers by income group and region

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Trained teachers'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ IncomeGroup$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Trained teachers by income group'$)+$

xlab$(\text{'}$Trained teachers $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Trained teachers'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ Region$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Trained teachers by region'$)+$

xlab$(\text{'}$Trained teachers $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

# Indicator: Literacy rate by income group and region

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Literacy rate'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ IncomeGroup$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Literacy rate by income group'$)+$

xlab$(\text{'}$Literacy rate $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Literacy rate'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ Region$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Literacy rate by region'$)+$

xlab$(\text{'}$Literacy rate $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

# Indicator: Unemployment by income group and region

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Unemployment'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ IncomeGroup$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Unemployment by income group'$)+$

xlab$(\text{'}$Unemployment $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$

education$\_$mf $\%>\%$

filter$($str$\_$detect$(`$Indicator Name$`,$ 'Unemployment'$))\%>\%$

ggplot$()+$

geom$\_$boxplot$($aes$($x $=$ value, fill $=$ Region$))+$

facet$\_$wrap$($ ~ male$\_$or$\_$female$)+$

ggtitle$(\text{'}$Unemployment by region'$)+$

xlab$(\text{'}$Unemployment $(\%)\text{'})+$

scale$\_$y$\_$continuous$($breaks $=$ NULL, labels $=$ NULL$)$