Shiny Web Applications

Advanced 15 hrs 1 Concepts

Your Learning Map

📌 You already know

You can compute results and build plots in R.

🎯 You'll learn here

Building an interactive web app with Shiny — reactive ui and server, inputs and outputs.

🌍 Where it's used

Dashboards anyone can use in a browser — rank predictors, NGO trackers, school ERPs — no R needed by the viewer.

🔗 Unlocks next

Ships your analysis as the interactive front-end of the capstone.

Shiny Architecture

Concept 1

UI and Server Basics

A Shiny app has two parts:

UI (User Interface): defines layout and input/output widgets
Server: the R function that computes outputs from inputs

library(shiny)

ui <- fluidPage(
  titlePanel('Student Score Explorer'),
  sidebarLayout(
    sidebarPanel(
      selectInput('subject', 'Subject:',
                  choices = c('Math','Science','English')),
      sliderInput('min_score', 'Min Score:', min=0, max=100, value=70),
      downloadButton('dl', 'Download Data')
    ),
    mainPanel(
      plotOutput('hist', height='350px'),
      tableOutput('summary_tbl')
    )
  )
)

server <- function(input, output, session){
  # Reactive expression — recalculates when inputs change
  filtered <- reactive({
    students |>
      filter(subject == input$subject, score >= input$min_score)
  })

  output$hist <- renderPlot({
    ggplot(filtered(), aes(x=score)) +
      geom_histogram(fill='#2563eb', binwidth=5) +
      labs(title=paste('Scores —', input$subject)) +
      theme_minimal()
  })

  output$summary_tbl <- renderTable({
    filtered() |> summarise(n=n(), mean=mean(score), sd=sd(score))
  })

  output$dl <- downloadHandler(
    filename = function() paste0(input$subject,'_data.csv'),
    content  = function(file) write_csv(filtered(), file)
  )
}

shinyApp(ui, server)

R — A minimal Shiny app

library(shiny)
ui <- fluidPage(
  sliderInput("n", "Sample size", 10, 100, 30),
  plotOutput("hist")
)
server <- function(input, output) {
  output$hist <- renderPlot(hist(rnorm(input$n)))
}
shinyApp(ui, server)

Output

(Shiny apps launch an interactive web page — run this in RStudio or deploy to shinyapps.io. It cannot run inside this lesson sandbox.)

Solved Examples

Example 1 Show a notification when the filtered data has fewer than 5 rows.

observe({
  if (nrow(filtered()) < 5){
    showNotification(
      paste('Only', nrow(filtered()), 'students match your filters.'),
      type = 'warning'
    )
  }
})
# Observers run whenever reactive dependencies change
# They're used for side effects (notifications, writing files)

Self-Assessment (2 questions)

Q1. What is the difference between reactive() and observe() in Shiny?

reactive() creates a cached value used in outputs. observe() runs code for its side effects (saving files, showing notifications) without returning a value.

Q2. What does input$subject refer to?

Shiny input objects are reactive. input$subject returns whatever the user has selected in the selectInput('subject', ...) widget.

Reproducible Research — R Markdown Data Ethics & Best Practices