HumanCapitalIndex.R

In ZellW/LearnPlotting:

# Data science, artificial intelligence, automation, and other advanced technologies are reshaping the world.
# # We're starting to see glimpses of it, for example the rapid emergence of self driving cars. Moreover, 
# the changes brought about by technology will likely become more frequent and more dramatic in the next few years.
# 
# In fact, two MIT economist, Erik Brynjolfsson and Andrew McAfee, think that the emergence of AI and data 
# driven technologies may be one of the most important events in all of human history. 
# 
# Risk and opportunity in the age of AI
# 
# They suggest that AI and automation will dramatically increase productivity and human development. 
# In essence, AI, machine learning, and data-driven technology will free us from our cognitive limitations 
# much like the engine freed us from our physical limitations. This will in turn lead to large gains 
# in productivity.
# 
# But while AI and data-driven tech will likely increase productivity, they are also likely to lead to 
# high levels of unemployment. As software and technology become more intelligent, they will replace people 
# who don't have the right skills.
# 
# A critical factor for success: human capital
# 
# How can someone succeed in such an era?
# 
# Brynjolfsson and McAfee point out that one of the keys to success in the age of data, AI, and automation will be human capital.
# 
# As technology increasingly substitutes for people, one of the keys to success will be cutting-edge knowledge, 
# judgement (particularly in complex domains), and creativity.
# 
# Mapping Human Capital
# 
# In thinking about this, I decided to map which countries have the highest levels of human capital, 
# as measured by the World Economic Forum.
# 
# While the WEF's Human Capital Index is not a perfect index of cutting edge skill, it does provide a high-level 
# view of the countries that may be better prepared for a changing, high-tech world.
# 
# One thing in particular that I'll note, is that this map effectively shows averages of national performance. 
# Having said that, talent and innovation are known to be highly uneven geographically. There are "spikes." 
# So even though this map shows countries' average performance, we'd likely see an even different story 
# if we looked at cities (or even zip codes).

#==============
# LOAD PACKAGES
#==============
library(tidyverse)
library(stringr)
library(readr)
library(rvest)
library(viridis)

#============
# SCRAPE DATA
#============

#-----------------------
# ESTABLISH URL LOCATION
#-----------------------
html.human_capital <- read_html("http://reports.weforum.org/human-capital-report-2016/rankings/")

#-------------------------------
# SCRAPE: Extract table from URL
#-------------------------------
df.human_capital <- html.human_capital %>% html_node("table") %>% html_table()

# inspect
glimpse(df.human_capital)

#========================
# RENAME VARIABLES
# - convert to lower case
#========================

df.human_capital <- df.human_capital %>% rename(rank = `Overall Rank`, country_nm = Economy
                    ,human_cap_score = `Overall Score`)

# INSPECT
glimpse(df.human_capital)

#==============
# GET WORLD MAP
#==============

map.world <- map_data("world")

# INSPECT
glimpse(map.world)

#===================================================================
# RECODE COUNTRY NAMES
#  some of the countries in the data frame df.human_capital don't 
#  exactly match the names in map.world.
#  This prohibits a proper join.  
#  We're going to hard-code (i.e. recode) these values to make sure
#  that the join works properly
#===================================================================

#------------------------------------------
# GET LIST OF COUNTRIES IN df.human_capital
#------------------------------------------
df.hcap_countries <- df.human_capital %>% select(country_nm)

#-----------------------------------
# GET LIST OF COUNTRIES IN map.world
#-----------------------------------
df.map_countries <- map.world %>% select(region) %>% group_by(region) %>% summarise(match = 1) 

#-----------------------------
# LOOK FOR MATCH or NO-MATCH
#  - we'll use a join for this
#    and inspect the output
#-----------------------------

df.check_match <- left_join(df.hcap_countries, df.map_countries, by = c('country_nm' = 'region'))

df.check_match %>% filter(is.na(match))

#------------------------------
# THESE ARE THE COUNTRIES THAT
# DO NOT MATCH:
#             country_nm  match
# 1   Russian Federation     NA
# 2       United Kingdom     NA
# 3        United States     NA
# 4           Korea Rep.     NA
# 5      Kyrgyz Republic     NA
# 6      Slovak Republic     NA
# 7        Macedonia FYR     NA
# 8    Iran Islamic Rep.     NA
# 9  Trinidad and Tobago     NA
# 10             Lao PDR     NA
# 11       CÙte d'Ivoire     NA
#------------------------------

#---------------------------------------------------------
# RECODE COUNTRY NAMES
#  - now that we have the names from df.human_capital that
#    don't match map.world, we can recode them
#  - to get the "new" names that we need to use,
#    inspect the names in df.map_countries
#---------------------------------------------------------
df.human_capital[123,2] <- "Ivory Coast"
df.human_capital$country_nm <- recode(df.human_capital$country_nm
                                      ,'Russian Federation'    = 'Russia'
                                      ,'United Kingdom'        = 'UK'
                                      ,'United States'         = 'USA'
                                      ,'Korea, Rep.'           = 'South Korea'
                                      ,'Kyrgyz Republic'       = 'Kyrgyzstan'
                                      ,'Slovak Republic'       = 'Slovakia'
                                      ,'Macedonia, FYR'        = 'Macedonia'
                                      ,'Iran, Islamic Rep.'    = 'Iran'
                                      ,'Trinidad and Tobago'   = 'Trinidad'
                                      ,'Lao PDR'               = 'Laos'
                                      )

#====================================================
# JOIN HUMAN CAPITAL DATA TO MAP
# - now we can join the human capital data to the map
#====================================================

map.world <- left_join(map.world, df.human_capital, by = c('region' = 'country_nm'))

glimpse(map.world)

#=====================================
# PLOT: BASIC MAP
# - this is a simple first iteration
#=====================================
ggplot(data = map.world, aes(x = long, y = lat, group = group)) + geom_polygon(aes(fill = human_cap_score)) 

#=======================================
# PLOT FINAL MAP
# - this is the finalized version after 
#   much iteration
#=======================================

ggplot(data = map.world, aes(x = long, y = lat, group = group)) +
     geom_polygon(aes(fill = human_cap_score)) +
     scale_fill_viridis(name = "Human Capital\nIndex") +
     labs(title = "Human Capital Index, by country"
          ,subtitle = "source: World Economic Forum, 2016\n reports.weforum.org/human-capital-report-2016/") +
     theme(panel.background = element_rect(fill = "#3E3E3E")
           ,plot.background = element_rect(fill = "#3E3E3E")
           ,legend.background = element_blank()
           ,axis.title = element_blank()
           ,axis.text = element_blank()
           ,axis.ticks = element_blank()
           ,panel.grid = element_blank()
           ,text = element_text(family = "Gill Sans", color = "#DDDDDD")
           ,plot.title = element_text(size = 32)
           ,legend.position = c(.18,.375)
     )