<h1>Interactively building test forms from an IRT perspective: An application of R and Shiny</h1>
<h2>Brandon LeBeau</h2>
<h3>University of Iowa</h3>
# Overview
<img src=" /figs/flowchart.png" alt=""/>
# 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
<img src="/figs/Rlogo.png" alt=""/>
# Reproducible Research
- Reproducible research has become popular.
- Commonly a document that contains both analysis and text.
- This can be done with `Rmarkdown` and `knitr.`
<img src="/figs/rmarkdown.PNG" alt=""/>
<img src="/figs/knitr.PNG" alt=""/>
# 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
<img src="/figs/shiny.PNG" alt=""/>
# Iterative Task Examples
- Building Assessments
- Exploratory Data Analysis
- Exploring Missing Data Patterns
- Model Selection/Building
# Iterative Analysis Structure
<img src="/figs/useless_meeting.PNG" alt=""/>
# What is Shiny?
- `Shiny` is an interactive web application framework for R.
+ Example: <http://shiny.rstudio.com/gallery/movie-explorer.html>
<img src="/figs/shiny_example.PNG" alt=""/>
# 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
# User Interface
- Simple user interface example from RStudio
- <http://shiny.rstudio.com/gallery/telephones-by-region.html>
```r
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")
)
)
)
)
```
# Server File
- The server file for RStudio example
- <http://shiny.rstudio.com/gallery/telephones-by-region.html>
```r
shinyServer(function(input, output) {
output$phonePlot <- renderPlot({
barplot(WorldPhones[ , input$region] * 1000,
main = input$region,
ylab = "Number of Telephones",
xlab = "Year")
})
})
```
# Interactivity is Key
<img src="/figs/interactivity.png" alt=""/>
# Tools for Interactivity
- Interactive Graphics
+ Using JavaScript - D3 graphics (`rCharts`)
+ Interactive static graphics - Garrett's presentation
- Interactive Tables
+ Using DT R package
# Reporting from Shiny
- Using `Rmarkdown` and `knitr` to create customizable reproducible reports
- Example: generate report button
- Generate final data files
- Example: download data button
# 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
# Weaknesses of Using Shiny
1. May take more time to develop initially
2. Need some R familiarity for development
# 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)
# 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
```
# Logistic Curve
<img src="/figs/logistic-1.png" alt=""/>
# Demo
<https://github.com/lebebr01/BuildForm>
```r
# Basic Theme
shiny::runGitHub('lebebr01/BuildForm', subdir = 'R', ref = 'basic')
# shinydashboard
shiny::runGitHub('lebebr01/BuildForm', subdir = 'R', ref = 'testmodule')
```
# 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
# Weaknesses of Shiny for Iterative Data Analysis
1. Need to train users
- Analysis
- Navigating web application
2. Knowledge of JavaScript, CSS, or HTML useful.
# 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.
# Summary
<img src="/figs/flowchart.png" alt=""/>
# 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/>
# Questions?
- Twitter: @blebeau11
- Website: <http://brandonlebeau.org>
- Slides: <http://brandonlebeau.org/2016/02/18/cspshiny/>