DATA210-dslabs-gapminder

#######
##DATA-2100
##Adefoluke Shemsu
#Week 2
#######

##------------------------------
#PROBLEM 1: EXPLORING GAPMINDER DATA

#1. Loading gapminder data from dslabs.

setwd("~/Documents/Education/Penn/Classes/DATA 210/Week 2")

install.packages("dslabs")
library(dslabs)
data(gapminder)
View(gapminder)

#2. Determining object type

class(gapminder)

#Object type = data.frame

#3. Checking gapminder data dimensions

dim(gapminder)

#10,545 rows, 9 columns

#4. Finding column names

colnames(gapminder)

# Names: Country, infant_mortality, fertility, gdp, region, year,
# life_expectancy, population, continent

#5. Finding hidden attributes:

attributes(gapminder)

#6. This dataframe likely lacks hidden attributes for the reason discussed during
# the lecture, which was that most .csv data has already been organized and defined
# appropriately based on what is available in that file type.

#7. Finding the population of Norway in 1962.
# Using the View() function:

View(gapminder)

# Population in 1962 = ~3.64 million people
# Finding Norway '62 by column/row:

subset(gapminder$country == "Norway", gapminder$year == 1962)

#8. Finding countries with life expectancy lower than 40.

lowlifeexpectancy <- subset(gapminder, gapminder$life_expectancy < 40) #To isolate < 40 life expectancy

#Finding countries with low life expectancy in the 2000s and later.

subset(lowlifeexpectancy, lowlifeexpectancy$year > 2000)

#Haiti and the Caribbean had life expectancies of less than 40 in the 21st century.

##--------------------------------
# PROBLEM 2: COLLEGE MAJORS, Part A

#1.Reading recent grads data.

library(readr)
recent_grads <- read_csv("~/Documents/Education/Penn/Classes/DATA 210/Week 2/recent-grads.csv")
View(recent_grads)

#2. Printing dimensions, class, row names, column names, etc.

dim(recent_grads)

# 173 rows, 21 columns

class(recent_grads)

# Object type = data.frame

# Getting names of rows/columns

colnames(recent_grads)
rownames(recent_grads)

# Checking for attributes

attributes(recent_grads)

# There are attributes for the columns.

##---------------------------------
# PROBLEM 2: COLLEGE MAJORS, Part B

#1. Finding the major category with the highest and lowest median incomes.

max(recent_grads$Median) #Verifying lowest and highest median incomes
min(recent_grads$Median) # Max = 110,000, Min = 22,000

#Highest = Engineering
#Lowest = Education

#2. Finding the avg median salary for both of those categories.

engineering <- subset(recent_grads, #Avg median for Engineering
                      recent_grads$Major_category == "Engineering")
mean(engineering$Median) #Avg median salary for engineers is $57,382.76

education <- subset(recent_grads,
                    recent_grads$Major_category == "Education")
mean(education$Median) #Avg median salary for education professionals is $32,350


#3. Drilling down to data that meets criteria for whether there are professions where
#(a) one of the top 10 professions with the highest median income and 
# (b) are in the top ten professions with the highest share of women.

majorityfemale <- subset(recent_grads, recent_grads$ShareWomen >= .51)
View(majorityfemale) 

# Of these professions, Astronomy and Astrophysics (ranked #8 of all jobs) both 
# had the highest median income among professions with > 50% share of women ($62,000), though
# it did not make the cut for being in the top 10 of professions dominated by them.

#4. Are women underrepresented in STEM fields?

View(majorityfemale)
View(recent_grads)

# Given that only 1 of the top 10 most lucrative professions are female "dominated"
# (53.5%), I would say that women are underrepresented specifically in Engineering rather than STEM,
# since there are professions aligned with STEM (e.g., Nursing) where women
# account for 89.6% of the workforce. The true under-representation, however, is found in what STEM 
# fields that appear to interest women more pay compared to the more computational
# and engineering-driven fields that appear to attract more men.