Intro to
Shiny

Day 23

Prof Amanda Luby

Carleton College
Stat 220 - Winter 2026

Demo

Shiny: High level view

Every Shiny app has a webpage that the user visits,
and behind this webpage there is a computer that serves this webpage by running R.

When running your app locally, the computer serving your app is your computer.

When your app is deployed, the computer serving your app is a web server.

Shiny vs. plotly

Shiny graphs need to be “connected” to RStudio or an Rstudio server
- Why? Data sets are rewrangled and a new graphic is drawn
plotly isn’t changing the underlying data set/stats being displayed, can be displayed on webpage without an RStudio session

Getting started

Option 1: Embed a shiny plot/table in HTML docs

Need add runtime: shiny to your YAML header

Option 2: Can create an app.R file with a ui() and server() function

For template
New Files > R Markdown > Shiny - New Files > Shiny Web App

What’s in an app?

library(shiny)
ui <- fluidPage(
  inputs,
  outputs
)


server <- function(
    <reactive functions and code to render output>) {
  ...
}
`

shinyApp(
  ui = ui, 
  server = server
)

User interface controls the layout and appearance of app
Server function contains instructions needed to build app

Defining inputs

ui <- fluidPage(
  inputPanel(
        selectInput('xcol', 
                    label = 'X Variable', 
                    choices = colnames(manager_survey), 
                    selected = names(manager_survey)[16]),
      )
)

inputPanel() creates an input list consisting of the user’s selections
selectInput() makes a dropdown select menu
xcol is the name of the input variable. Make sure input names are unique!

Render output

server <- function(input, output) {
    output$scatterPlot <- renderPlot({
    manager_survey |>
      filter(industry %in% input$industry) |>
      ggplot( aes_string(x = input$xcol, y = "annual_salary")) +
      geom_jitter(size = 2, alpha = 0.6, aes(color = industry)) +
      theme(legend.position = "top")
  })
}

Access input values with input$input_name
Save objects to display to output$object_name
Build objects to display with renderSomething()

Add output to the UI function

ui <- fluidPage(
      inputPanel(
        selectInput('xcol', 
                    label = 'X Variable', 
                    choices = colnames(manager_survey), 
                    selected = names(manager_survey)[16]),
      ),
    
   plotOutput(outputId = "scatterPlot")
)

typeofOutput("object_name") will try to pull off the object_name element of the output list
Make sure output names are unique

Add a title

ui <- fluidPage(
  
    titlePanel(title = "Exploring the Ask a Manager data"),

      inputPanel(
        selectInput('xcol', 
                    label = 'X Variable', 
                    choices = colnames(manager_survey), 
                    selected = names(manager_survey)[16]),
      ),
    
   plotOutput(outputId = "scatterPlot")
)

typeofOutput("object_name") will try to pull off the object_name element of the output list
Make sure output names are unique

Data: Ask a manager

Source: Ask a Manager Survey via TidyTuesday

This data does not reflect the general population; it reflects Ask a Manager readers who self-selected to respond, which is a very different group (as you can see just from the demographic breakdown below, which is very white and very female).

Some findings here.

Data: `manager-survey.csv`

manager <- read_csv("https://stat220-w26.github.io/data/manager-survey.csv")
manager

# A tibble: 26,232 × 18
   timestamp          how_old_are_you industry  job_title additional_context_o…¹
   <chr>              <chr>           <chr>     <chr>     <chr>                 
 1 4/27/2021 11:02:10 25-34           Educatio… Research… <NA>                  
 2 4/27/2021 11:02:22 25-34           Computin… Change &… <NA>                  
 3 4/27/2021 11:02:38 25-34           Accounti… Marketin… <NA>                  
 4 4/27/2021 11:02:41 25-34           Nonprofi… Program … <NA>                  
 5 4/27/2021 11:02:42 25-34           Accounti… Accounti… <NA>                  
 6 4/27/2021 11:02:46 25-34           Educatio… Scholarl… <NA>                  
 7 4/27/2021 11:02:51 25-34           Publishi… Publishi… <NA>                  
 8 4/27/2021 11:03:00 25-34           Educatio… Librarian High school, FT       
 9 4/27/2021 11:03:01 45-54           Computin… Systems … Data developer/ETL De…
10 4/27/2021 11:03:02 35-44           Accounti… Senior A… <NA>                  
# ℹ 26,222 more rows
# ℹ abbreviated name: ¹additional_context_on_job_title
# ℹ 13 more variables: annual_salary <dbl>, other_monetary_comp <dbl>,
#   currency <chr>, currency_other <chr>, additional_context_on_income <chr>,
#   country <chr>, state <chr>, city <chr>,
#   overall_years_of_professional_experience <chr>,
#   years_of_experience_in_field <chr>, …

Your turn

Run the app! What does it show us?
Read the code: find the inputs and the outputs
Adapt the app to allow the user to specify what variable maps to the point color in the scatterplot
Adapt the app to allow the user to filter the data by annual_salary. Allow values between the minimum and maximum annual_salary selected.
Adapt the app to allow the user to select whether the plot should have a white background
Add a histogram that allows the user to select what quantitative variable should be displayed. Add any other interactive elements (e.g., binwidth adjustments) that you want to.

Intro to Shiny