Load the required libraries
library(tidyverse)
library(plotly)
library(shiny)Read in the data
cdc_mortality_df <- read.csv('https://raw.githubusercontent.com/charleyferrari/CUNY_DATA_608/master/module3/data/cleaned-cdc-mortality-1999-2010-2.csv')
cdc_mortality_df <- as_tibble(cdc_mortality_df)Check the head
head(cdc_mortality_df)## # A tibble: 6 x 6
## ICD.Chapter State Year Deaths Population Crude.Rate
## <chr> <chr> <int> <int> <int> <dbl>
## 1 Certain infectious and parasitic dis~ AL 1999 1092 4430141 24.6
## 2 Certain infectious and parasitic dis~ AL 2000 1188 4447100 26.7
## 3 Certain infectious and parasitic dis~ AL 2001 1211 4467634 27.1
## 4 Certain infectious and parasitic dis~ AL 2002 1215 4480089 27.1
## 5 Certain infectious and parasitic dis~ AL 2003 1350 4503491 30
## 6 Certain infectious and parasitic dis~ AL 2004 1251 4530729 27.6Check the summary
summary(cdc_mortality_df)## ICD.Chapter State Year Deaths
## Length:9961 Length:9961 Min. :1999 Min. : 10
## Class :character Class :character 1st Qu.:2002 1st Qu.: 177
## Mode :character Mode :character Median :2005 Median : 667
## Mean :2005 Mean : 2929
## 3rd Qu.:2008 3rd Qu.: 2474
## Max. :2010 Max. :96511
## Population Crude.Rate
## Min. : 491780 Min. : 0.00
## 1st Qu.: 1728292 1st Qu.: 4.60
## Median : 4219239 Median : 24.00
## Mean : 5937896 Mean : 52.15
## 3rd Qu.: 6562231 3rd Qu.: 50.50
## Max. :37253956 Max. :478.40# rename the ICD.Chapter to ICD_chapter and Crude.Rate to Crude_Rate
cdc_mortality_df <- cdc_mortality_df %>% rename(ICD_Chapter = ICD.Chapter, Crude_Rate = Crude.Rate)
cdc_mortality_df## # A tibble: 9,961 x 6
## ICD_Chapter State Year Deaths Population Crude_Rate
## <chr> <chr> <int> <int> <int> <dbl>
## 1 Certain infectious and parasitic di~ AL 1999 1092 4430141 24.6
## 2 Certain infectious and parasitic di~ AL 2000 1188 4447100 26.7
## 3 Certain infectious and parasitic di~ AL 2001 1211 4467634 27.1
## 4 Certain infectious and parasitic di~ AL 2002 1215 4480089 27.1
## 5 Certain infectious and parasitic di~ AL 2003 1350 4503491 30
## 6 Certain infectious and parasitic di~ AL 2004 1251 4530729 27.6
## 7 Certain infectious and parasitic di~ AL 2005 1303 4569805 28.5
## 8 Certain infectious and parasitic di~ AL 2006 1312 4628981 28.3
## 9 Certain infectious and parasitic di~ AL 2007 1241 4672840 26.6
## 10 Certain infectious and parasitic di~ AL 2008 1385 4718206 29.4
## # ... with 9,951 more rowsdf = cdc_mortality_df
states <- unique(df$State)
df_summary <- df %>%
group_by(Year,ICD_Chapter) %>%
summarise(State = "Average",
Deaths = mean(Deaths),
Population = mean(Population),
Crude_Rate = 100000 * sum(Deaths)/sum(Population)) %>%
bind_rows(df)## `summarise()` has grouped output by 'Year'. You can override using the `.groups`
## argument.df2 <- df_summaryAs a researcher, you frequently compare mortality rates from particular causes across different States. You need a visualization that will let you see (for 2010 only) the crude mortality rate, across all States, from one cause (for example, Neoplasms, which are effectively cancers). Create a visualization that allows you to rank States by crude mortality for each cause of death.
#Filter the data for only 2010 and from one cause say Neoplasms
cdc_mortality_df_2010 = cdc_mortality_df %>% filter(Year == 2010 & ICD_Chapter == 'Neoplasms') %>%
arrange(State, desc(Crude_Rate))
head(cdc_mortality_df_2010)## # A tibble: 6 x 6
## ICD_Chapter State Year Deaths Population Crude_Rate
## <chr> <chr> <int> <int> <int> <dbl>
## 1 Neoplasms AK 2010 910 710231 128.
## 2 Neoplasms AL 2010 10429 4779736 218.
## 3 Neoplasms AR 2010 6616 2915918 227.
## 4 Neoplasms AZ 2010 11129 6392017 174.
## 5 Neoplasms CA 2010 57820 37253956 155.
## 6 Neoplasms CO 2010 7220 5029196 144.ShinyApp for Question1
# Build the UI for question1
ui <- fluidPage(
# Panel for Question 1
tabsetPanel(
tabPanel(title = "Question1",
sidebarPanel(
htmlOutput('message_q1'),
sliderInput('year_q1','Year',min(df$Year),max(df$Year),2010,1,width=600,
sep = ""),
selectInput('causes_q1', 'Causes',
unique(df$ICD_Chapter), selected='Neoplasms',width = 600)
),
mainPanel(plotlyOutput('plot1_q1'))
)
))
# build the server for question1
server <- function(input, output, session) {
# Question1 Server Side
data_q1 <- reactive({
df2 <- df %>%
filter(Year == input$year_q1) %>%
filter(ICD_Chapter == input$causes_q1) %>%
arrange(Crude_Rate)
df2$State <- factor(df2$State, levels = unique(df2$State))
df2
})
# output message for question1
output$message_q1 <- renderText({
df2 <- data_q1()
if (nrow(df2) == 0) {
paste("<h4><center>No Data for<b>",input$causes_q1,
"</b>in",input$year_q1,"</center></h4>", sep =" ")
}
else {
paste("<h4><center>The Mortality Rates for<b>",input$causes_q1,
"</b>in",input$year_q1,"<br>(scaled by multiplying 100000)</center></h4>", sep =" ")
}
})
# output plot for question1
output$plot1_q1 <- renderPlotly({
df2 <- data_q1()
if (nrow(df2) == 0) {
plotly_empty(type = "bar",width = 1,height = 1,)
}
else {
plot_ly(x = df2$Crude_Rate,
y = df2$State,
orientation='h',
width = 900,
height = 750,
type = "bar") %>%
layout(
yaxis = list(tickfont = list(size = 11)))
}
}
)
}
shinyApp(ui = ui, server = server)## PhantomJS not found. You can install it with webshot::install_phantomjs(). If it is installed, please make sure the phantomjs executable can be found via the PATH variable.Often you are asked whether particular States are improving their mortality rates (per cause) faster than, or slower than, the national average. Create a visualization that lets your clients see this for themselves for one cause of death at the time. Keep in mind that the national average should be weighted by the national population.
# Build the UI part of the shinyApp
ui <- fluidPage(
tabsetPanel(
# Panel for Question 2
tabPanel(title = "Question2",
sidebarPanel(
htmlOutput('message_q2'),
selectInput('causes_q2', 'Causes',
unique(df$ICD_Chapter), selected='Neoplasms',width = 600),
checkboxGroupInput("States_q2", "States",states,inline = TRUE)
),
mainPanel(plotlyOutput('plot1_q2'))
)
)
)
# Build the server side of the shinyApp
server <- function(input, output, session) {
#Question2 Server Side
data_q2 <- reactive({
df3 <- df2 %>%
filter(ICD_Chapter == input$causes_q2)
df3
})
state2_q2 <-reactive(
input$States_q2
)
# output message for question2
output$message_q2 <- renderText({
paste("<h4><center>The Mortality Rates for<b>",input$causes_q2,
"</b><br>(scaled by multiplying 100000) </center></h4>", sep =" ")
})
# output plot for question2
output$plot1_q2 <- renderPlotly({
df3 <- data_q2()
if (nrow(df3) == 0) {
plotly_empty(type = "scatter",width = 1,height = 1,)
}
else {
temp <- filter(df3, State == "Average")
fig <- plot_ly(
x = temp$Year,
y = temp$Crude_Rate,
mode = 'lines+markers',
name = "Average",
type = "scatter"
)
selected_states <- state2_q2()
for (state in states)
{
alpha <- ifelse(state %in% selected_states,1,0.05)
temp <- filter(df3, State == state)
fig <- fig %>% add_trace(x = temp$Year, y = temp$Crude_Rate, name = state, mode = 'lines+markers',opacity = alpha)
}
fig
}
})
}
shinyApp(ui = ui, server = server)# Build the UI part of the shinyApp
ui <- fluidPage(
# Panel for Question 1
tabsetPanel(
tabPanel(title = "Question1",
sidebarPanel(
htmlOutput('message_q1'),
sliderInput('year_q1','Year',min(df$Year),max(df$Year),2010,1,width=600,
sep = ""),
selectInput('causes_q1', 'Causes',
unique(df$ICD_Chapter), selected='Neoplasms',width = 600)
),
mainPanel(plotlyOutput('plot1_q1'))
),
#---------------------------------------------------------------------------
# Panel for Question 2
tabPanel(title = "Question2",
sidebarPanel(
htmlOutput('message_q2'),
selectInput('causes_q2', 'Causes',
unique(df$ICD_Chapter), selected='Neoplasms',width = 600),
checkboxGroupInput("States_q2", "States",states,inline = TRUE)
),
mainPanel(plotlyOutput('plot1_q2'))
)
)
)
# Build the server side of the shinyApp
server <- function(input, output, session) {
# Question1 Server Side
data_q1 <- reactive({
df2 <- df %>%
filter(Year == input$year_q1) %>%
filter(ICD_Chapter == input$causes_q1) %>%
arrange(Crude_Rate)
df2$State <- factor(df2$State, levels = unique(df2$State))
df2
})
# output message for question1
output$message_q1 <- renderText({
df2 <- data_q1()
if (nrow(df2) == 0) {
paste("<h4><center>No Data for<b>",input$causes_q1,
"</b>in",input$year_q1,"</center></h4>", sep =" ")
}
else {
paste("<h4><center>The Mortality Rates for<b>",input$causes_q1,
"</b>in",input$year_q1,"<br>(scaled by multiplying 100000)</center></h4>", sep =" ")
}
})
# output plot for question1
output$plot1_q1 <- renderPlotly({
df2 <- data_q1()
if (nrow(df2) == 0) {
plotly_empty(type = "bar",width = 1,height = 1,)
}
else {
plot_ly(x = df2$Crude_Rate,
y = df2$State,
orientation='h',
width = 900,
height = 750,
type = "bar") %>%
layout(
yaxis = list(tickfont = list(size = 11)))
}
})
#-------------------------------------------------------------------------------
#Question2 Server Side
data_q2 <- reactive({
df3 <- df2 %>%
filter(ICD_Chapter == input$causes_q2)
df3
})
state2_q2 <-reactive(
input$States_q2
)
# output message for question2
output$message_q2 <- renderText({
paste("<h4><center>The Mortality Rates for<b>",input$causes_q2,
"</b><br>(scaled by multiplying 100000) </center></h4>", sep =" ")
})
# output plot for question2
output$plot1_q2 <- renderPlotly({
df3 <- data_q2()
if (nrow(df3) == 0) {
plotly_empty(type = "scatter",width = 1,height = 1,)
}
else {
temp <- filter(df3, State == "Average")
fig <- plot_ly(
x = temp$Year,
y = temp$Crude_Rate,
mode = 'lines+markers',
name = "Average",
type = "scatter"
)
selected_states <- state2_q2()
for (state in states)
{
alpha <- ifelse(state %in% selected_states,1,0.05)
temp <- filter(df3, State == state)
fig <- fig %>% add_trace(x = temp$Year, y = temp$Crude_Rate, name = state, mode = 'lines+markers',opacity = alpha)
}
fig
}
})
}
shinyApp(ui = ui, server = server)