< section>

Interactively building test forms from an IRT perspective: An application of R and Shiny

Brandon LeBeau

University of Iowa

< section>

Overview

< section>

R

• R is an open source statistical programming language.
  • Pros:
    • Common statistical procedures are found in R
    • Can extend functionality with packages/functions
  • Cons:
    • Need to be comfortable with code

< section>

Reproducible Research

• Reproducible research has become popular.
  • Commonly a document that contains both analysis and text.
  • This can be done with Rmarkdown and knitr.

< section>

Iterative/Interactive Data Analysis

• This type of analysis requires some input from the user.
  • Data analysts may use R
  • Shiny is a great option for code novices

< section>

Iterative Task Examples

• Building Assessments
• Exploratory Data Analysis
• Exploring Missing Data Patterns
• Model Selection/Building

< section>

Iterative Analysis Structure

< section>

What is Shiny?

• Shiny is an interactive web application framework for R.
  • Example: http://shiny.rstudio.com/gallery/movie-explorer.html

< section>

Components of Shiny

1. User Interface (ui.r)
   • What the user sees and interacts with
2. R Analysis (server.r)
   • The R code running behind the scenes

< section>

User Interface

• Simple user interface example from RStudio
  • http://shiny.rstudio.com/gallery/telephones-by-region.html

shinyUI(

  fluidPage(    

    titlePanel("Telephones by region"),

    sidebarLayout(      

      sidebarPanel(
        selectInput("region", "Region:", 
                    choices = colnames(WorldPhones)),
        hr(),
        helpText("Data from AT&T (1961) The World's Telephones.")
      ),

      mainPanel(
        plotOutput("phonePlot")  
      )

    )
  )
)

< section>

Server File

• The server file for RStudio example
  • http://shiny.rstudio.com/gallery/telephones-by-region.html

shinyServer(function(input, output) {

  output$phonePlot <- renderPlot({

    barplot(WorldPhones[ , input$region] * 1000, 
            main = input$region,
            ylab = "Number of Telephones",
            xlab = "Year")
  })
})

< section>

Interactivity is Key

< section>

Tools for Interactivity

• Interactive Graphics
  • Using JavaScript – D3 graphics (rCharts)
  • Interactive static graphics – Garrett’s presentation
• Interactive Tables
  • Using DT R package

< section>

Reporting from Shiny

• Using Rmarkdown and knitr to create customizable reproducible reports
  • Example: generate report button
• Generate final data files
  • Example: download data button

< section>

Strengths of Using Shiny

1. The app can be written solely using R code
   • Can use CSS, JavaScript, or HTML as needed
2. User does not need to know any R
3. Many hosting options
4. Application can be as simple or complex as needed (both visually and functionally)
5. Flexible output

< section>

Weaknesses of Using Shiny

1. May take more time to develop initially
2. Need some R familiarity for development

< section>

Background for Demo

• In educational assessment, we need to create new test forms
  • Exposure concerns
  • Add new content
  • Altering test landscape
• Building test forms is an iterative process that involves gathering information from:
  • Item analyses
  • Test blueprints
  • Item response theory (IRT)

< section>

IRT Data

##      Item.1 Item.2 Item.3 Item.4 Item.5 Item.6 Item.7 Item.8
## [1,]      1      1      1      1      1      1      1      1
## [2,]      0      1      0      0      1      0      1      0
## [3,]      1      1      1      0      1      0      1      0
## [4,]      0      1      0      1      1      0      1      0
## [5,]      0      1      1      1      1      0      1      1
## [6,]      1      1      0      0      1      0      1      0

< section>

Logistic Curve

< section>

Demo

https://github.com/lebebr01/BuildForm

# Basic Theme
shiny::runGitHub('lebebr01/BuildForm', subdir = 'R', ref = 'basic')

# shinydashboard
shiny::runGitHub('lebebr01/BuildForm', subdir = 'R', ref = 'testmodule')

< section>

Benefits of Shiny for Iterative Data Analysis

1. Free valuable data analyst/scientist resources.
2. Improve data literacy in the organization.
3. Highly customizable
   • Analysis (server.r)
   • User interface (ui.r)
   • Reporting

< section>

Weaknesses of Shiny for Iterative Data Analysis

1. Need to train users
   • Analysis
   • Navigating web application
2. Knowledge of JavaScript, CSS, or HTML useful.

< section>

Guidelines for Building Shiny Apps

1. Understand reactive coding.
2. Modularize your code – define functions for repetitive code chunks.
3. Define scope early.
   • Define output.
4. Clean up UI last.

< section>

Summary

< section>

Shiny Resources

• http://shiny.rstudio.com/
• http://shiny.rstudio.com/articles/
• http://shiny.rstudio.com/gallery/
• https://www.rstudio.com/products/shiny/shiny-user-showcase/

< section>

Questions?

• Twitter: @blebeau11
• Website: http://educate-r.org
• Slides: http://educate-r.org/2016/02/18/cspshiny/