app.R

library(shiny)
library(wbstats)
library(tidyr)
library(ggplot2)
library(maps)
library(DT)
library(shinythemes)

#install.packages("shinythemes")

source("project.R")
#source("ui.R")
#source("server.R")




home <- tabPanel(
  "Home",  
  h1("Study on Economic Status and Education Rates"),
  p("Through this project, we are attempting to draw relationships and comparisons between 
   these two variables; economy rates of countries and higher rates of education in different countries.
   Some of the questions that will facilitate in drawing these comparisons between the two variables inlcude:",
   
  tags$ul(
    tags$li("Is there a relationship between economic status of a country and their graduation rates?",
       tags$ul(
         tags$li("This question is aimed to help us finally analyze and answer if there is a relationship 
         between the two factors; if there is a correlation between the two, and if so, the type of 
                 correlation (positive, negative).")
       )),
    tags$li("   Rank the education rate from highest to lowest among countries worldwide, and also show their 
            economic trend. How do these compare?", 
       tags$ul(
         tags$li("This question will help us understand which country has the highest rates of highest education 
                 and which has the lowest rates.")
       )),
    tags$li("How does the correlation of US higher education rates vs. economy look like with respect to US events?",
       tags$ul(
         tags$li("This question will help us understand how these values compare to how well the country is faring.")
       )),
    
    tags$li("Which regions tend to have a higher GDP? Higher graduation rates? What could these results entail?",
       tags$ul(
         tags$li("This will help us understand if there is a certain region where more affluent individuals use overseas bank accounts to store money, for example. These findings could prompt many more questions.")
         )
       )
    )
  ), 
  p("We chose this topic, as we believe that education is a very essential tool to a person's overall success in life irrespective of age. Just like every human being requires oxygen to survive, 
   education too is very essential to survive in this world. Education provides people the knowledge and skills they need in order to survive. 
   By analyzing these questions relating to education and how the economy rates of countries affect education - we hope to be able to better 
   understand which countries are doing well in terms of economy and higher rates of education, and which countries are not doing well in these aspects. 
   By understanding these aspects, we as students can work on spreading awareness regarding countries that are low on higher education rates and work in small 
   ways that impact those countries and improve the overall higher educational rates. Thus, we believe that the topic chosen is important."), 

   h3("Data Description"),
   p("Our data is from the World Bank and the Organisation for Economic Co-Operation and Development. Our World Bank data shows countries' GDP per capita and the 
   OECD data shows education rates for different countries. OECD has columns of education rates for upper-secondary education, post-secondary education, 
   short-cycle tertiary education, as well as college degrees such as a bachelor's or equivalent. We are focusing on people who obtain bachelor's degrees the first time they enter university.
   The data was collected and distributed by the respective organizations."),
  
  p(strong(a("World Bank Data", href = "https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?end=2018&start=2018&view=bar"))),
  
  p(strong(a("OECD Data", href = "https://stats.oecd.org/Index.aspx?datasetcode=EAG_GRAD_ENTR_RATES"))),
  
  p(strong(a("Project Report", href = "https://info201a-wi20.github.io/project-report-Malia-T/"))),
  
  h3("Team Members"),
  p(
    tags$ul(
      tags$li("Malia Cortez"),
      tags$li("Serah Prakkat"),
      tags$li("Ryan Ros"),
      tags$li("Mei Zhao")
    ),
    class = "team-paragraph"
  )
)

#Question 1 Tab#
################
q1 <- tabPanel(
  "Econ. Status & Grad Rates",
  titlePanel("Is there a relationship between economic 
             status and graduation rates?"),
  sidebarLayout(
    sidebarPanel(
      p("The Economy and Rates of education change graph shows a positive relationship. 
      It shows that as the GDP in USD increases, the relative graduation rates also 
      increase. This means that as the economy of the country does well, the more the 
      percentage of people are able to attend and complete their college degrees."),
      plot1_input_world <- selectInput(inputId = "year_select_plot1", label = "Year",
                                       choices = c(2005, 2010, 2011, 2012, 2013, 2014, 
                                                   2015, 2016, 2017), 
                                       selected = 2005),
      
      plot1_input_country <- selectInput(inputId = "country_select_plot1", label = "Country",
                                         choices = mean_data$Country, 
                                         selected = "Argentina"),
      p("From this plot, we can say that it is apparent that there is a positive correlation between mean GDP and mean graduation rates worldwide. 
      It is important to note that there are significant outliers in the data that have showed up on the plot. 
      In the case of Argentina, the trend line seems to be showing a positive relation; as GDP increases, the graduation rate increases. 
      A strong correlation between mean GDP and mean rate of graduation would suggest more devastating consequences for developing countries, or countries with a lower GDP who are not as economically successful. 
      A cycle in which countries that are lower in GDP additionally have lower college education rates is doomed to continue in such a pattern. Of course, this data is not accounting for how many individuals go 
        abroad to achieve a higher education, so perhaps that could be worth looking into as well.")
    ),
    mainPanel(
      tabsetPanel(
        tabPanel("Average", plotOutput(outputId = "plot_1_mean")),
        tabPanel("Worldwide", plotOutput(outputId = "plot_1_output_worldwide")),
        tabPanel("By Country", 
                 p(plotOutput(outputId = "eco_trend")),
                 p(plotOutput(outputId = "edu_trend")))
      ))
  )
)

#Question 2 Tab#
################
q2 <- tabPanel(
 "Education & Econ. Status Ranked",
  titlePanel("Education Rate Rankings Among Countries Worldwide and Their Economic Trend."),
  sidebarLayout(
    sidebarPanel(  
      sliderInput(inputId = "country_slider", label = "Select The Number Of Countries To Show",min = 1, max = 40, value = 10) # Creates a slider input to select the number of countries to display on the bar graph
    ),
    mainPanel(
      h3("Comparing the Ranks of Education Rate from Highest to Lowest Among Countries Worldwide and Their Economic Trend"), # Heading level 3
      p("What we can  gather from this visualization and quantitative data is that there seems to be no outright correlation between a country's economic status and a country's rates of people pursuing higher education.
     We saw a bit of this earlier with Luxembourg's perplexing case; the country had one of the lowest graduation rates but the largest GDP.
     When comparing average education rates and average GDP, there seems to be little to no relationship between the two.
     All in all, perhaps there is a better way of measuring economic success rather than simply said country's GDP, and it may also just be dependent on the country itself.
     In order to find a relationship between these two general areas, we could look at country income, infrastructure, and other measures of economic status to perform more analysis.
     Given the vast range of the data, it is possible that the slight correlation we saw in the last visualization is simply due to that same range, as well as outliers.
     From this visualization, however, we can see that although we reordered our GDP in decreasing order, there is not as much of a distinct pattern in the respective graduation rates.
     Within the context of the previous visualization, this plot highlights how much it matters to have variety. Upon first glance, there seems to be little to no relationship in this data, however the relationship is quite strong as we saw in the previous plot.
     If we wanted to ask more questions, we could look at the data on a country-by-country basis over time so we can analyze those findings as well."),
      p(plotOutput(outputId = "eco_bar_plot")), # Displays eco_bar_plot bar chart on main panel
      p(plotOutput(outputId = "edu_bar_plot")), # Displays edu_bar_plot bar chart on main panel
      p(tableOutput("mean_data")) # Displays mean_data data frame table on side bar panel
    ), position = "left", fluid = TRUE
  )
)

#Question 3 Tab#
################
q3 <- tabPanel(
  "US Correlation & US Events",
  titlePanel("Comparing Higher Education Rates and Economy based on US Events"),
  p("Here, we are attempting to see how US economy and graduation rates are affected by certain US events.
    We can gather from this visualization that US GDP was not impacted very much in respect to these events listed in the table. 
    We can also see that graduation rate suffered a considerable dip in growth from 2011 - 2014. 
    When looking at these findings, we can see that this may have been due to the US Fiscal Cliff.	
    From [Wikipedia](https://en.wikipedia.org/wiki/United_States_fiscal_cliff): The United States fiscal cliff was a situation that took place in January 2013 when several previously-enacted laws came into effect simultaneously, increasing taxes and decreasing spending. 
    Since these taxes were increased so much, perhaps many families could not afford to send their children to college, or perhaps they encouraged their child to go to trade school or pursue another option instead. Families also may not have been able to afford to pay for resources and help so that their children had a viable path to college. We found it very interesting to find that the country's GDP was not affected considering an event such as a fiscal cliff in which the name suggests economic consequence. 
    This visualization shows that the people are going to be much more negatively affected by certain events than US economic status ever will be."),
	
  sidebarLayout(
    sidebarPanel(
      radioButtons(inputId = "events_select", label = "Select Certain United States Event to Filter",
                   choices = list("All Events",
                                  "Patient Protection and Affordable Care Act, Dodd-Frank Wall Street Reform and Consumer Protection Act",
                                  "Japan Tohoku earthquake and tsunami", 
                                  "U.S. Fiscal cliff",
                                  "Budget sequestration",
                                  "Quantitative easing (QE) ends (aka large-scale asset purchases)", 
                                  "Trans-Pacific Partnership, Joint Comprehensive Plan of Action (aka Iran nuclear deal)", 
                                  "Presidential race",
                                  "Trump Tax Act (Tax Cuts and Jobs Act)"), 
                   selected = "All Events")
    ),
    mainPanel(
      h3("How does the correlation of US higher education rates vs. economy look with respect to US events?"), # Heading level 3
      tabsetPanel(
        tabPanel(
          "Gross Domestic Product (GDP)",
          plotOutput(outputId = "event_gdp") # Displays event_gdp line chart on main panel
        ),
        tabPanel(
          "Graduation Rate",
          plotOutput(outputId = "event_grad") # Displays event_grad line chart on main panel
        ),
        tabPanel(
          "GDP & Graduation Rates",
          p(plotOutput(outputId = "event_gdp_tab3")), # Displays event_gdp line chart on main panel
          p(plotOutput(outputId = "event_grad_tab3")) # Displays event_grad line chart on main panel
        )
      ),
      p(tableOutput("usa")) # Displays usa data frame table on side bar panel
    )
  )
)


#Question 4 Tab#
################
q4 <- tabPanel(
  "Worldwide GDP & Graduation Rate",
  #select year#
  #############
  
  year_input <- selectInput(
    inputId = "year_map",
    label = "Year",
    choices = c(2005,2010,2011,2012,2013,2014,2015,2016,2017),
    selected = 2005),
  mainPanel(
    tabsetPanel(
      tabPanel("Education", 
               sidebarLayout(
                 mainPanel(plotOutput(outputId = "edu_map_plot")),
                 sidebarPanel(strong(textOutput("mean_world_edu")),tableOutput("mean_edu_data")),
                 position = "left"
               )),
      tabPanel("Economy", 
               sidebarLayout(
                 mainPanel(plotOutput(outputId = "eco_map_plot")),
                 sidebarPanel(strong(textOutput("mean_world_eco")),tableOutput("mean_eco_data")),
                 position = "left"
               ))),
    p("This visualization suggests that the countries with the highest graduation rate tend to be clustered around Scandinavia and Western Europe, Japan, Australia and New Zealand all have very high education rates as well. These same countries generally all had very high GDPs. Particularly this cluster around Scandinavia and Western Europe is of interest.
    One aspect that is apparent here is the amount of missing data that is on the map. Nearly all of Africa is missing besides South Africa. Much of South America is missing, much of the Middle East, and many Eastern European countries have no data. Given this fact, we cannot draw too many conclusions without worlwide comprehensive data. 
    That being said, we can discuss this cluster of success in Scandinavia and Western Europe. [This article](https://www.investopedia.com/articles/managing-wealth/042916/offshore-banking-isnt-illegal-hiding-it.asp) from Investopedia suggests that many very affluent people may keep foreign money in developed nations in offshore bank accounts/tax havens. 
    This could have an impact on country GDP and actually contribute positively to it; overseas bank accounts can give account holders opportunities to internationally invest, contributing to the foreign country's GDP. As we saw previously, there is actually a strong correlation, outliers aside, between GDP and graduation rates, so this would also make 
      sense for explaining why those countries' graduation rates are high as well.")
  )
)


##############################################################

ui <- fluidPage (
  theme = shinytheme("slate"),
  includeCSS("style.css"),
  navbarPage(
    title = "Info 201, H4",
    home,
    q1,
    q2,
    q3,
    q4
  )
)



##############################################################

server <- function(input, output) {
  
  #edu map#
  #########
  
  output$edu_map_plot <- renderPlot({
    year_input <- input$year_map
    edu_input <- df %>% 
      filter(Year == as.integer(year_input))
    where_to_cut <- c(-Inf, 0, 10, 20, 30, 40, 50, 60, Inf)
    label <- c("0%","0% to 10%", "10% to 20%", "20% to 30%", "30% to 40%", "40% to 50%", "50% to 60%","> 60%")
    edu_df <- mutate(edu_input, change_labels = cut(edu_input$Education, breaks = where_to_cut, labels = label))
    grad_map_df <- left_join(world_map, edu_df, by = "iso3c")
    grad_map_df$change_labels <- factor(grad_map_df$change_labels, levels = rev(levels(grad_map_df$change_labels)))
    q4_edu_map <- ggplot(data = grad_map_df, mapping = aes(x = long, y = lat)) +
      geom_polygon(aes(group = group, fill = change_labels)) +
      scale_fill_brewer(palette = "BuPu",direction = -1) +
      coord_quickmap() +
      labs(title = "Worldwide Graduation Rate") +
      theme_void()+
      guides(fill = guide_legend(title = "Graduation Rate"))
    return(q4_edu_map)
  })
  #eco map#
  ######### 
  output$eco_map_plot <- renderPlot({
    year_input <- input$year_map
    eco_input <- df %>% 
      filter(Year == as.integer(year_input))
    where_to_cut <- c(-Inf,0 , 10000, 25000, 40000, 55000, 70000, 85000, Inf)
    label <- c("< 10000%","$10,000 to $25,000", "$25,000 to $40,0000", "$40,000 to $55,000", "$55,000 to $70,000", "$70,000 to $85,000", "$85,000 to $100,000","> $100,000")
    eco_df <- mutate(eco_input, change_labels = cut(eco_input$Economy, breaks = where_to_cut, labels = label))
    eco_map_df <- left_join(world_map, eco_df, by = "iso3c")
    eco_map_df$change_labels <- factor(eco_map_df$change_labels, levels = rev(levels(eco_map_df$change_labels)))
    q4_eco_map <- ggplot(data = eco_map_df, mapping = aes(x = long, y = lat)) +
      geom_polygon(aes(group = group, fill = change_labels)) +
      scale_fill_brewer(palette = "OrRd",direction = -1) +
      coord_quickmap() +
      labs(title = "Worldwide Economy GDP Rate") +
      theme_void()+
      guides(fill = guide_legend(title = "GDP Rate"))
    return(q4_eco_map)
  })
  
  
  #eco & edu relationship trend#
  ##############################   
  output$plot_1_output_worldwide <- renderPlot({
    # used for plotting worldwide over years
    plot1_input_world <- input$year_select_plot1
    plot_1_year_input <- filter(df, Year == as.integer(plot1_input_world))
    
    q1_plot_world <- ggplot(data = plot_1_year_input, aes(x = Economy, y = Education ))+
      
      geom_point(color = "red")+
      labs(title = "Economy and Rates of Education Change", x = "GDP in USD", y = "Graduation Rate %")+
      geom_smooth(method=lm, se = FALSE)+
      theme_minimal()+
      theme(axis.line = element_line(size = 0.5, linetype = "solid",colour = "black"))
    return(q1_plot_world)
  })
  
  
  #eco bar chart Question 2#
  ########################## 
  output$eco_bar_plot <- renderPlot({
    mean_data %>%
      arrange(-mean_data$mean_gdp) %>%
      head(input$country_slider) %>%
      ggplot(aes(x = reorder(Country, -mean_gdp), y = mean_gdp))+
      geom_col(fill = "orange")+
      labs(y = "GDP in US Dollars",
           title = "Ranking of Average Education Rate and GDP Between Countries"
      )+
      theme_minimal()+
      theme(axis.title.x=element_blank(),
            axis.text.x=element_blank(),
            axis.title.y = element_text(size = 12),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 2)),
            axis.line.y = element_line(size = 0.5, linetype = "solid",
                                       colour = "black")
      )
    
  })
  
  #edu bar chart Question 2#
  ##########################   
  output$edu_bar_plot <- renderPlot({
    mean_data %>%
      arrange(-mean_data$mean_gdp) %>%
      head(input$country_slider) %>%
      ggplot(aes(x = reorder(Country, -mean_gdp), y = mean_grad_rate))+
      geom_col(fill = "pink")+
      labs( x = "Country", 
            y = "Graduation Rate")+
      theme_minimal()+
      theme(axis.text.x = element_text(angle = 90, size = 12,hjust = 1, vjust = 0.25),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 18)),
            legend.position = "none",
            axis.title = element_text(size = 12),
            axis.line = element_line(size = 0.5, linetype = "solid",
                                     colour = "black")
            
      )    
    
  })
  
  
  #Mean Data Table for Question 2#
  ################################
  output$mean_data <- renderTable({
    mean_data %>%
      arrange(-mean_data$mean_gdp) %>%
      head(input$country_slider)
    
  })
  
  #event gdp chart for Question 3#
  ################################
  output$event_gdp <- renderPlot({
    usa %>%
      filter(Events == input$events_select | (input$events_select == "All Events")) %>%
      ggplot(aes(x = Year,y = Economy)) +
      geom_point(size = 3)+
      geom_line()+
      labs(y = "GDP in US Dollars")+
      scale_x_continuous(breaks=seq(2010, 2017, 1))+
      theme_minimal()+
      theme(axis.title = element_text(size = 12),
            axis.text.x = element_text( size = 12),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 18)),
            axis.line = element_line(size = 0.5, linetype = "solid",
                                     colour = "black")
      )
  })
  
  output$event_gdp_tab3 <- renderPlot({
    usa %>%
      filter(Events == input$events_select | (input$events_select == "All Events")) %>%
      ggplot(aes(x = Year,y = Economy)) +
      geom_point(size = 3)+
      geom_line()+
      labs(y = "GDP in US Dollars")+
      scale_x_continuous(breaks=seq(2010, 2017, 1))+
      theme_minimal()+
      theme(axis.title = element_text(size = 12),
            axis.text.x = element_text( size = 12),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 18)),
            axis.line = element_line(size = 0.5, linetype = "solid",
                                     colour = "black")
      )
  })
  
  #event grad chart for Question 3#
  #################################
  output$event_grad <- renderPlot({
    usa %>%
      filter(Events == input$events_select | (input$events_select == "All Events")) %>%
      ggplot(aes(x = Year,y = Education)) +
      geom_point(size = 3)+
      geom_line()+
      labs(y = "Graduation Rate %")+
      scale_x_continuous(breaks=seq(2010, 2017, 1))+
      theme_minimal()+
      theme(axis.title = element_text(size = 12),
            axis.text.x = element_text( size = 12),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 18)),
            axis.line = element_line(size = 0.5, linetype = "solid",
                                     colour = "black")
      )
  })
  
  output$event_grad_tab3 <- renderPlot({
    usa %>%
      filter(Events == input$events_select | (input$events_select == "All Events")) %>%
      ggplot(aes(x = Year,y = Education)) +
      geom_point(size = 3)+
      geom_line()+
      labs(y = "Graduation Rate %")+
      scale_x_continuous(breaks=seq(2010, 2017, 1))+
      theme_minimal()+
      theme(axis.title = element_text(size = 12),
            axis.text.x = element_text( size = 12),
            axis.text.y = element_text(margin = margin(t = 0, r = 0, b = 0, l = 18)),
            axis.line = element_line(size = 0.5, linetype = "solid",
                                     colour = "black")
      )
  })
  
  #Dynamic usa Table for Question 3#
  ########################################
  output$usa <- renderTable({
    usa %>%
      filter(Events == input$events_select | (input$events_select == "All Events"))
  })
  
  #table for q4 education
  output$mean_edu_data <- renderTable({
    df %>%
      filter(Year == input$year_map) %>%
      select(Country,Education) %>% 
      arrange(-Education)
    
  })
  
  #table for q4 economy
  output$mean_eco_data <- renderTable({
    df %>%
      filter(Year == input$year_map) %>%
      select(Country,Economy) %>% 
      arrange(-Economy)
    
  })
  
  #text analysis for q4 edu
  output$mean_world_edu <- renderText({
    mean_grad <- world_mean %>% 
      filter(Year == input$year_map) %>% 
      pull(mean_grad_rate)
    paste("The world average graduation rate in",input$year_map,"is",round(mean_grad, digits = 2),"%.")
  })
  
  #text analysis for q4 eco
  output$mean_world_eco <- renderText({
    mean_gdp <- world_mean %>% 
      filter(Year == input$year_map) %>% 
      pull(mean_gdp)
    paste("The world average GDP (in US Dollars) in",input$year_map,"is $",round(mean_gdp, digits = 2),".")
  })
  
  # mean plot for q1
  output$plot_1_mean <- renderPlot({
    mean_plot <- ggplot(mean_data,aes(x = mean_gdp,y = mean_grad_rate ))+
      geom_point(color = "red")+
      labs(title = "Economy and Rates of Education Change", x = "GDP in USD", y = "Graduation Rate %")+
      geom_smooth(method=lm, se = FALSE)+
      theme_minimal()+
      theme(axis.line = element_line(size = 0.5, linetype = "solid",colour = "black"))
    
    return(mean_plot)
  })
  
  # Used for country output: gdp
  output$eco_trend <- renderPlot({
    plot1_input_country <- input$country_select_plot1
    plot1_countries <- df %>%
      filter(Country == plot1_input_country)

    q1_plot_eco <- ggplot(data = plot1_countries, aes(x = Year, y = Economy)) +
      geom_point(color = "red") +
      labs(title = paste("Economy for", plot1_input_country, "Over Time"), x = "Years", y = "GDP in USD") +
      xlim(2005, 2017) +
      geom_smooth(method = lm, se = FALSE) +
      theme_minimal() +
      theme(axis.line = element_line(size = 0.5, linetype = "solid", colour = "black"))
    return(q1_plot_eco)
  })

  # Used for country output: edu
  output$edu_trend <- renderPlot({
    plot1_input_country <- input$country_select_plot1
    plot1_countries <- df %>%
      filter(Country == plot1_input_country)

    q1_plot_edu <- ggplot(data = plot1_countries, aes(x = Year, y = Education)) +
      geom_point(color = "red") +
      labs(title = paste("Education for", plot1_input_country, "Over Time"), x = "Years", y = "Education Rate") +
      xlim(2005, 2017) +
      geom_smooth(method = lm, se = FALSE) +
      theme_minimal() +
      theme(axis.line = element_line(size = 0.5, linetype = "solid", colour = "black"))
    return(q1_plot_edu)
  })
  
}

shinyApp(ui = ui, server = server)