App Overview

This Shiny app enables the user to analyze fate of the RMS Titanic’s passengers & crew. via exploratory plotting of the passenger survival data.

The data used by the App comes from one of R’s built in datasets –Titanic.  But it can very easily be modified to accept other dataset & plot them.   
It downloads the dataset & converts it to a dataframe so that it can be later passed on to the FILTER command from the DPLYR package

library(shiny)
library(shinythemes)
library(dplyr, quietly = TRUE , warn.conflicts = FALSE)
library(ggplot2, quietly = TRUE)
library(readr, quietly = TRUE)

data("Titanic")
Titanic <- as.data.frame(Titanic)

The dataset has 5 variables and 32 observations:

App Layout

• The App has a SIDEBAR Panel that has 2 sliders where the user will input the visualization choices
• The user can choose the visualise all people that were onboard of the ship, OR by passenger category
  • 1st class passengers
  • 2nd class passengers
  • 3rd class passengers
  • ship’s crew
• PLUS the user can choose if they want to visualize either gender at one given time, or both  
• The MAIN Panel displays the plot of the user selection
• It also displays the sum of the total # of passengers contained in the user selection

R code and Shiny commands used for the App

The UI code accepts User Input via 2 “selectInput” Shiny commands:
* PassengerClass
* Sex

selectInput(inputId = "PassengerClass", label = "Enter passenger category",
                        choices = c(levels(Titanic$Class), "All")  ),
                        
            selectInput("Sex", "Choose sex", choices = c(levels(Titanic$Sex), "Both")   ),
            submitButton(text = "Show Plot")

The Shiny SELECT boxes were improved in their functionality, so that they can also accept an input of ALL categories of the PASSENGER CLASS & SEX variables from the dataset.    

The algorithm in the server.R file used to subset the Titanic dataset (based on the user selection) is provided below:

ifelse (input$PassengerClass =="All", Titanic_subset <- Titanic, 
                     Titanic_subset <- Titanic %>% filter(Class == input$PassengerClass)
                     )
              
              if (input$Sex == "Male") {Titanic_subset <- Titanic_subset %>% filter(Sex == "Male")}
              if (input$Sex == "Female") {Titanic_subset <- Titanic_subset %>% filter(Sex == "Female")}
              if (input$Sex == "Both") {Titanic_subset <- Titanic_subset}
              
              ploteo <- ggplot(Titanic_subset, aes(x = Class, y = Freq, fill = Survived ) ) + 
                               geom_bar(stat = "identity", position = position_dodge() ) +
                               ylab("Number of passengers")
                              
              ploteo

Exploratory analysis

By making exploratory plots, we can quickly obtain insights from this dataset.

For example, the overall survival for 2nd Class passenger tells us that 1/3 of the passengers survived, but if we differentiate by sex we will see that the % of females that survived was roughly inversely proportional to the male survivors. Thus, we can confirm that the “women first” call was indeed observed for the 2nd Class passengers.

ALL 2nd Class passengers
User input PASSENGERCLASS = “All”
User input SEX = “Both”

## [[1]]
## [1] 2nd 2nd 2nd 2nd 2nd 2nd 2nd 2nd
## Levels: 1st 2nd 3rd Crew

2nd Class FEMALE passengers

## [[1]]
## [1] 2nd 2nd 2nd 2nd 2nd 2nd 2nd 2nd
## Levels: 1st 2nd 3rd Crew

2nd Class MALE passengers

## [[1]]
## [1] 2nd 2nd 2nd 2nd 2nd 2nd 2nd 2nd
## Levels: 1st 2nd 3rd Crew