Louisville Salaries Dashboard.Rmd

---
title: "Louisville Metro Government Salary Explorer"
output: 
  flexdashboard::flex_dashboard:
    theme: yeti
    social: menu
    source: embed
runtime: shiny
---

```{r setup, include=FALSE}
# Load Packages
library(flexdashboard)
library(dplyr)
library(ggvis)
library(shiny)
library(ggplot2)
library(plotly)
library(knitr)
library(scales)

# GGplot Theme
theme_set(theme_gray(base_size=13))

# Pull Data
df <- read.delim(url("http://api.louisvilleky.gov/api/File/DownloadFile?fileName=SalaryData.txt"),stringsAsFactors = F)
df$CalendarYear <- factor(df$CalendarYear)
df$JobTitle <- factor(trimws(df$JobTitle))
df$Department <- factor(trimws(df$Department))
df$EmployeeName <- factor(df$EmployeeName)


```


Inputs {.sidebar}
=========================================
This application pulls the latest salary data from the <a href="http://portal.louisvilleky.gov/service/data">Louisville Open Data Portal</a>, which provides data from 2008 to present.

Plots are interactive, hover for information. Further options are available in the top right of each plot window.

```{r}
selectizeInput("Departments","Tab 2: Select Departments to Overlay",levels(df$Department),selected=NULL,multiple=TRUE)
selectInput("Department","Tab 3: Pick Single Department",levels(df$Department))
selectInput("Jobtitle","Tab 4: Select a job title",choices=levels(df$JobTitle),selected="Cashier")
selectizeInput("Employee","Tab 5: Pick Employee",levels(df$EmployeeName), selected="Adams, William",options = list(maxOptions=20000))

```

1) Total Salaries/Employees
=========================================

Column
---------------------------------------
### The sum of all annual salaries by year

```{r}
totals <- df %>%
  group_by(CalendarYear) %>%
  summarise("Total" = sum(AnnualRate))

renderPlotly({
  ggplot(totals, aes(x=CalendarYear,y=Total,group=1))+
    geom_point(col="blue")+
    geom_line(col="blue") +
    labs(x=NULL,y=NULL,title=NULL)+
    scale_y_continuous(labels = comma)

})
```

### Number of employees by year

```{r}
emps <- df %>%
  group_by(CalendarYear) %>%
  summarise("Total Employees" = n())

renderPlotly({
  ggplot(emps, aes(x=CalendarYear, y=`Total Employees`, group=1))+
    geom_point(col="blue")+
    geom_line(col="blue")+
    labs(x=NULL,y=NULL,title=NULL)
})
```

Column
--------------------------

### Employees by Department (2016)

```{r}
bydept <- df %>%
  group_by(CalendarYear,Department) %>%
  summarise(Employees = n()) %>%
  filter(CalendarYear == 2016)

bydept$Department <- reorder(bydept$Department,bydept$Employees)

renderPlotly({
  ggplot(bydept, aes(x=Department,y=Employees))+
           geom_bar(stat="identity", fill="blue")+
           coord_flip()+
           labs(x=NULL,y=NULL,title=NULL)
})

```


2) Across Depts
=========================================
### Select departments to overlay from the sidebar (type to search) for a comparison of salary distributions.


```{r}
dens <- reactive({
  df <- df[df$Department %in% input$Departments,]
})

renderPlotly({
  if(is.null(input$Departments)){
    return(NULL)
  }
  ggplot(dens(), aes(x=AnnualRate, fill=Department))+
  geom_density(alpha=.6)+
  scale_x_continuous(labels = comma)+
  labs(x="Yearly Salary",y="Density",title=NULL)

})
```

3) Within Depts
=========================================
### Select a department from the sidebar to see the top ten salaries from that department for all available years. 

```{r}
highest <- reactive({
  df %>%
  filter(Department == input$Department) %>%
  group_by(CalendarYear) %>%
  arrange(desc(AnnualRate)) %>% 
  top_n(10)
})
  
renderPlotly({
  ggplot(highest(), aes(x=EmployeeName,y=AnnualRate))+
    geom_bar(stat="identity",fill="blue") +
    facet_wrap(~CalendarYear,scales="free_x") +
    scale_y_continuous(labels = comma)+
    labs(y="Salary",x="Employee")+
    theme(axis.text.x = element_text(angle = 90, hjust = 1))

})  
  



```

4) Job Titles
=========================================
### Select any job title from the sidebar to see the salary distribution (all years) by department.

```{r}
titles <- reactive({
  df %>%
    filter(JobTitle == input$Jobtitle)
})

renderPlotly({
  ggplot(titles(), aes(x=AnnualRate, fill=Department))+
    geom_histogram(col="white",position="identity",alpha=.6)+
    scale_x_continuous(labels = comma)+
    labs(x="Yearly Salary",y="Count",title=paste(input$Jobtitle,"Income by Department",sep=" "))
})
```


5) By Employee {data-orientation=columns}
=====================================     

Column
-------------------------
### Select any employee from the sidebar (type to search).


```{r}
employee <- reactive({
    df2 <- df %>%
    filter(EmployeeName == input$Employee) %>%
    arrange(desc(CalendarYear))

})

renderPlotly({
  ggplot(employee(),aes(x=CalendarYear,y=AnnualRate,group=1))+
    geom_point(col="blue")+
    geom_line(col="blue")+
    labs(x=NULL,y="Yearly Salary",title=paste(input$Employee,"Salary Trend"))+
    scale_y_continuous(labels = comma)
})    
```

```{r}
dep <- renderText({
  as.character(employee()[1,3])
})

jobt <- renderText({
  as.character(employee()[1,4])
})

wages <- renderText({
  paste("$",as.character(format(employee()[1,10],big.mark=",")),sep="")
})

minsal <- renderText({
  paste("$",as.character(format(min(employee()[,5]),big.mark=",")),sep="")
})

maxsal <- renderText({
  paste("$",as.character(format(max(employee()[,5]),big.mark=",")),sep="")
})

overtime <- renderText({
  paste("$",as.character(format(sum(employee()[,7]),big.mark=",")),sep="")
})

```

Column {data-width=200}
--------------------------
### Employee Information

This employee is in the <b>`r dep`</b> department. Their job title is <b>`r jobt`</b>. So far in 2016, this employee has received <l style="color:green">`r wages`</l> in compensation. 

Their lowest salary in the available data is <l style="color:green">`r minsal`</l> and their highest salary in the available data is <l style="color:green">`r maxsal`</l>.

This employee has received <l style="color:green">`r overtime`</l> in overtime pay since 2008.