---
title: "Activity: Spatial Data"
editor: source
---

## Instructions

For this activity, work in groups of 2 - 3. Work together to come to a solution, and help each other out when stuck! The goal is to use the journey as a vessel for your and your peer's learning, not to make it to the 'correct' answer as fast as possible.

1.  Compose your answer to each problem between the bars of red stars.

2.  If partially completed code is provided, you will see that `#| eval: FALSE` in the `R` chunk options. After editing/completing the code, make sure to change that to `eval = TRUE`.

3.  Be prepared to ask for help from me and your classmates!

## Problem 0: loading Libraries

Load the following libraries. You may need to install them first if you haven't used them on your computer before.

```{r libraries}
# Libraries
library(tidyverse)
library(tidycensus)
library(sf)
```

## Problem 1: Setting up Census API Key

Like we did for Data Intake Activity 1, we are going to use Census data from the API package `tidycensus`:

```{r}
library(tidycensus)
```

Type your API key into the quotes in the code provided below, and change `#| eval: true` in the `R` chunk options. If you need a key, you can get one [here](https://api.census.gov/data/key_signup.html).

------------------------------------------------------------------------

```{r}
#| eval: false

# Load in your Census API key 
api_key <- "Insert your API key here"
```

------------------------------------------------------------------------

## Problem 2: Retrieving Spatial Data for Multnomah County, OR

Run the code below that uses the `get_acs()` command to bring in Oregon tract-level median household income, total foreign-born population, and total population from the 5-year 2018-2022 American Community Survey (ACS). Remember that “E” at the end of the variable indicates “Estimate” and “M” indicates margin of errors:

```{r}
# Use tidycensus to get ACS data for Multnomah County, OR
or_tracts <- get_acs(geography = "tract",
                     year = 2022,
                     variables = c(medincome = "B19013_001",
                                   fb = "B05012_003",
                                   totp = "B05012_001"),
                     state = "OR",
                     county = "Multnomah",
                     survey = "acs5",
                     output = "wide",
                     geometry = TRUE,
                     key = api_key)

# View the first few rows of the data
head(or_tracts)
```

Check the type of the `or_tracts` object using the `class()` function to confirm that it is both a data frame, and an `sf` object. What is the name of the variable in the data set that contains the spatial information? How did we tell the `get_acs()` function above to pull this information? (Hint: check the options for the `get_acs()` call in the code above.)

------------------------------------------------------------------------

------------------------------------------------------------------------

## Problem 3: Wrangling Data

Before we get to mapping, run the following code to (i) drop the margin of error columns, (ii) rename variables, and (iii) calculate a percent foreign born (`pfb`) variable:

```{r}
# A bit of data wrangling
or_tracts <- or_tracts %>%
  select(-ends_with("M")) %>% # getting rid of Margin of Error columns
  rename(medincome = medincomeE, fb = fbE, totp = totpE) %>% # renaming columns
  mutate(pfb = 100 * (fb / totp)) # creating % foreign born

# View the updated data
head(or_tracts)
```

Next, make a new variable called `medincome_quintiles`, a categorical variable which splits the the `medincome` variable into quintiles:

- Q1: \<=20th percentile

- Q2: \>20th percentile and \<=40th percentile

- Q3: \>40th percentile and \<=60th percentile

- Q4: \>60th percentile and \<=80th percentile

- Q5: \>80th percentile

------------------------------------------------------------------------

------------------------------------------------------------------------

## Problem 4: Mapping Data

Now that you’ve got your spatial data in and wrangled, the next step is to map something. Using the `sf` package, make a plot of Multnomah County, with tracts colored by their quintile for median income.

------------------------------------------------------------------------

------------------------------------------------------------------------

## Problem 5: Customization

Now that you've made your first spatial plot, it's time to beautify it! Change the fill color and clean up the visual properties of the plot. Feel free to add a base layer if you're feeling extra excited about spatial data!

------------------------------------------------------------------------

------------------------------------------------------------------------

## Problem 6: Adding Reed College

Reed College is about at a Longitude of -122.6306, and a Latitude of 45.48194. This dataset has these coordinates:

```{r}
locations <- data.frame(long = -122.6306, lat = 45.48194, what = "Reed")
```

Add Reed College to your plot as a point on the graph. Label it as Reed College, and make sure to choose a color that will stick out from the color scheme and other visual properties you chose in the previous question.

------------------------------------------------------------------------

------------------------------------------------------------------------

## Problem 7: Saving Maps

You can save your maps using the function `ggsave()`, for example:

```{r, eval=F}
ggsave(plot=OBJECT_NAME, filename="[FILE_NAME].[EXT]")
```

Define your map from the previous question as an object (with the name of your choice), and edit the above code, specifying:

- The name of your graph object (`OBJECT_NAME`)

- The file name for your graph (`[FILE_NAME]`)

- The extension or type of file you want the graph saved as (`[EXT]`). Try either pdf, png, or jpg.

------------------------------------------------------------------------

------------------------------------------------------------------------