---
title: "Exercise 5"
author: "Put your name here"
date: "Put the date here"
output: html_document
---

# Task 1: Reflection

Put your reflection here


# Task 2: Essential pandemic construction

The New York City Department of Buildings (DOB) maintains [a list of construction sites](https://www1.nyc.gov/assets/buildings/html/essential-active-construction.html) that have been categorized as "essential" during the city's shelter-in-place pandemic order.

## Load and clean data

First we load and clean the data (downloaded in May 2020):

```{r load-clean-data, warning=FALSE, message=FALSE}
# You'll only need the tidyverse library for this exercise
library(tidyverse)

# Load original data
essential_raw <- read_csv("data/EssentialConstruction.csv")

# Clean the data a little
# Some of the borough names are in ALL CAPS, so we use str_to_title() to convert
# everything in the column to title case.
# We also make BOROUGH and CATEGORY factors (or categorical variables)
essential <- essential_raw %>% 
  mutate(BOROUGH = str_to_title(BOROUGH),
         BOROUGH = factor(BOROUGH),
         CATEGORY = factor(CATEGORY))
```

## Approved projects by borough

Right now there's a row for each approved construction site. We need to condense that down to get counts of construction sites by different variables. We can do this by using `group_by()` and `summarize()`

```{r summarize-data-borough}
essential_by_borough <- essential %>% 
  group_by(BOROUGH) %>% 
  summarize(total = n()) %>% 
  mutate(proportion = total / sum(total))
```

```{r plot-borough-summary}
# Add plot with geom_col() here
```


## Approved projects by category

```{r summarize-data-category}
# Create a summarized dataset of projects by category
#
# I won't give you the code for this (big hint though: copy the code for the
# borough summary and change just one thing)
```

```{r plot-category-summary}
# Add a lollipop chart here
```


## Approved projects across borough and category

```{r summarize-data-heatmap}
# Create a summarized dataset of projects by both borough and category
#
# I also won't give you the code to make the summary for the heatmap. You'll
# need to group by two variables to make the summary. IMPORTANTLY you'll also
# need to add another group_by() in between summarize() and mutate(), otherwise,
# R will calculate percentages in unexpected groups. 
#
# If you want the percentages of categories to add up to 100% in each borough,
# you'll want to group by borough before calculating the proportion; if you want
# the percentages of boroughs to add up to 100% in each category, you'll want to
# group by category
```

```{r plot-heatmap}
# Add a heatmap here with geom_tile()
```