PADP 7120 Data Applications in PA

class: center, middle, inverse, title-slide

.title[
# PADP 7120 Data Applications in PA
]
.subtitle[
## RLab 2: Importing & Exporting
]
.author[
### Alex Combs
]
.institute[
### UGA | SPIA | PADP
]
.date[
### Last updated: January 16, 2024
]

---

# Objectives

- Import data

- included in a R package
  - csv and Excel

- Export R Markdown

- change YAML to print nicer tables
  - to Word and PDF
  - suppress code and/or results

- Export data to csv

---
# Set-up

> **Create a new project named "rlab2". I recommend locating it in the same parent file as your rlab1 project.**

---
# Set-up

> **Start a R Markdown document**

> **Change the title in the YAML to "RLab2: Import & Export"**

> **Keep the setup code chunk at the top of the template but delete the rest of the template.**

---
# Set-up: packages

> **In the setup code chunk at the top, add the below code. Run this code to load packages.**

```r
library(tidyverse) #includes function for importing CSV
library(readxl) #for importing Excel files
library(gapminder) #contains data we need to use
library(fivethirtyeight) #contains data we need to use
```

- If you receive error message "package does not exist," you need to install it.

---
# Set-up: files

> **Download "ncbirths" from eLC and add it to your current project folder**

> **Download "ga_schdist_raw" from eLC and add to your "rlab2" project folder.**

---
class: inverse, middle, center

# Data imports

---
# Data in R Packages

- Many R packages, like `gapminder` and `fivethirtyeight` include datasets

- We can browse datasets by going to the "Packages" tab in the bottom-right pane of RStudio
  - Click on the package that contains the relevant dataset
  - A list of datasets will appear
  - Click on a dataset to learn more about it

---
# Data in R Packages

> **In a new code chunk, add and run the following code**

```r
slice_head(gapminder, n=4)
# slice_head() prints row(s) from the top of a dataset
# n=4 tells it to print the top 4 rows
```

|country     |continent | year| lifeExp|      pop| gdpPercap|
|:-----------|:---------|----:|-------:|--------:|---------:|
|Afghanistan |Asia      | 1952|  28.801|  8425333|  779.4453|
|Afghanistan |Asia      | 1957|  30.332|  9240934|  820.8530|
|Afghanistan |Asia      | 1962|  31.997| 10267083|  853.1007|
|Afghanistan |Asia      | 1967|  34.020| 11537966|  836.1971|

- This is one way to provide a preview of the data

---
# Data in R Packages

> **Change the code so it saves as an object named `gap_preview` like so and run again**

```r
gap_preview <- slice_head(gapminder, n=4)
```

- You should now see an object named `gap_preview` in your environment (top-right pane)

---
# Printing the result

- Note that the table does not print when you run the code

- By default, R does not print the result when saved as an object

> **Type `gap_preview` on a separate line so the table prints when knit**

```r
gap_preview <- slice_head(gapminder, n=4)

*gap_preview
```

---
# Data in R Packages

- There is a dataset named `hate_crimes` contained within the `fivethirtyeight` package.

> **Insert a _new code chunk_ and add the following code**

```r
hc <- hate_crimes
```

- Now we have a copy of `hate_crimes` in our environment

---
# Data in R Packages

> **Start a new line in the code chunk you just created.**

> **Add code that will print the first six rows of the `hc` dataset.**

> **Knit to HTML**

---
class: inverse, middle, center

# Printing tables

---
# Printing tables

- We have made two tables that will print when we knit using the code:

```r
gap_preview # prints saved object in environment
slice_head(hc, n=4) # prints result without saving anything
```

- There is enough horizontal space for all `gap_preview` variables to display but not for all `hc` variables

---
# Printing tables

- We can improve how all tables print when knit by changing the YAML

> **Change the YAML to the example below. Make sure you indent correctly.**

```r
---
title: "RLab2: Import & Export"
author: "Your Name"
output:
* html_document:
*   df_print: kable
---
```

> **Knit to HTML**

- `gap_preview` table looks better but `hc` table runs off the page

---
# Printing tables

> **Change `df_print` option in YAML to paged like highlighted below:**

```r
---
title: "RLab2: Import & Export"
author: "Alex Combs"
output:
  html_document:
*   df_print: paged
---
```

> **Knit to HTML**

- Now we can interact with the `hc` table to view multiple pages of variables.

---
# YAML themes

This is not required, but you can change the theme of your document (similar to themed templates in Microsoft Office) like below.

```r
---
title: "RLab2: Import & Export"
author: "Alex Combs"
output:
  html_document:
    df_print: paged
*   theme: readable
---
```

-If interested, more info on [themes](https://bootswatch.com/3/) and [document formatting](https://bookdown.org/yihui/rmarkdown/html-document.html#appearance-and-style)

---
class: inverse, middle, center

# Code chunk options

---
# Code chunk options

- Can control how code knits using code chunk options

- Include options by adding a comma after the `r`, followed by a space

```r
{r, options_go_here}
```

- echo = TRUE/FALSE 
  - Suppresses the code from the output but shows results

```r
{r, echo=FALSE}
```

- include = TRUE/FALSE
  - Suppress the code AND all results from the output

```r
{r, include=FALSE}
```

---
# Code chunk options

> **Find the code chunk that prints the `gap_preview` table**

> **Add the code chunk option that suppresses the code AND results**

> **Find the code chunk that prints the first four rows of `hc`**

> **Add the code chunk option that suppresses ONLY the code**

> **Knit to HTML**

---
# Output options overview

![](labs_files/output_options.png)

- The minus signs indicate what each option suppresses

---
# Global code chunk option

- When you start a new Rmd template, it contains a code chunk near the top named "setup" that contains

```r
{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

- This sets `echo=TRUE` for all code chunks by default

> **Change to `echo=FALSE`, to suppress all code when knit.**

> **Add `message=FALSE` and `warning=FALSE` to suppress the background information RStudio sometimes prints but should not be included in most reports**

---
class: inverse, middle, center

# Back to data imports

---
# Importing data

- Organizations publish data files in a variety of formats that may be inconvenient but tend to use a consistent format across files and time

- Provides the opportunity to use code to import many similar files quickly

- Messy data files present a choice in workflow
  - Clean the data some as part of the import, or
  - Import data as is and clean later

- Avoid changing the data files in something like Excel before importing it because you or others can't replicate that as easily

---
# Importing CSV

- Generic syntax:

```r
any_name <- read_csv("filename.csv")
```

- Tells R to import the `"filename.csv"`, naming it whatever we type on the left side

- Remember:
  - Quotes around the file
  - Include the file extension .csv
  
> **In the setup code chunk, import ncbirths.csv naming the saved data as ncbirths.**

---
# Importing Excel

- Generic syntax:

```r
anyname <- read_excel("filename.xlsx")
```

- Works with both .xls or .xlsx file extensions

> **In the setup code chunk, import ga_schdist_raw.xlsx naming the saved data as ga_schdist_raw**

---
# Preserve or Clean on Import

- The previous two slides cover all the code needed to import the two most common public data file formats.

- If the data you want to import looks relatively clean, additional code is not neeeded.

- If data are messy, we may want to clean the data at the same time as importing it.

> **Click on the two data sets you just saved to the environment. Let's examine whether data look correct.**

---
# Importing data

- Starting with `ncbirths`

- Some desirable features of this dataset:
  - Variable names are in the top row
  - Variable names are short

- This dataset doesn't need changes

---
# Importing data

- Now viewing `ga_schdist_raw`, the data did not import as we want

- Can help to first open messy data in spreasheet software like Excel to understand what needs to be done

- Data has some undesirable qualities
  - At least first 6 rows should be skipped
  - Variable names are very long
  - Row 231 and below contain information we don't want to import
  - Note that the data we want to import runs from cells A7 to I230

---
# Importing data

- Common cleaning tasks to combine with importing:

- Specifying which rows and columns to import if the raw file has non-data information in it
  - Changing how R will store a variable

- We can handle these tasks using RStudio's import function

---
# Importing data

> **In file pane, click on "ga_schdist_raw.xlsx", then "Import Dataset..."**

- Can click on a variable to change how R stores it, or exclude it from the import

- Can skip some specified number of the top rows, or specify a range of cells

- Can specify whether the top row includes variable names

---
# Importing data

- The RStudio import window in the bottom-right provides the code needed to import the data based on the options we choose

- **Important:** You must paste this code in your Rmd or else the data won't import when you knit

---
# Importing data

> **Notice there is a "Range" option in the bottom left. Enter `A7:I230`.**

> **There are numeric variables stored as character. Click on each and change to numeric using the dropdown menu.**

> **Copy code and paste in the _setup code chunk_**

---
# Importing data

- We need to change the code we just pasted in a few ways:

> **Delete the part of the file path unique to your computer and include only the file name.**

- This is important for sharing documents. Others will encounter an error when trying to run code that contains a file path unique to your computer.

> **Delete the last line containing View because it often causes an error when knit**

> **Run this code to complete import.**

---
# Importing data

- Variable names are long but this dataset is ready to use

- Do not *have* to rename variables but typing long variables can be inconvenient

- What if we wanted to rename these variables to make it easier to work with

---
# Renaming variables

Generic syntax

```r
dataset <- rename(dataset, new_name = current_name, new_name = "current name",...)
```

- If a current variable name has a space, need to use quotes

---
# Renaming variables

- Let's rename one variable in `ga_schdist_raw`

> Refer to the example code on previous slide to rename the first variable, "Agency Name", to district. Add this code to your setup code chunk.

> Run this code to check whether it works

---
class: inverse, center, middle

# Exporting Document and Data

---
# Knit to Word

- If Word is installed on your computer, knitting to Word should work

- Good way to get all text, graphs, and tables in a final Word document

- Can then change style and formatting

---
# Knit to PDF

- To knit to PDF, you can install a light version of LaTeX within RStudio.

```r
# In the console pane, type
install.packages('tinytex') # hit Enter
# Next, in the console pane, type
tinytex::install_tinytex() # hit Enter
```

---
# Export data to CSV

- We may want to export data in our environment that we have prepared in RStudio for others to use

- Generic syntax for exporting data to CSV:

```r
write_csv(name_of_object_in_environment, 'name_for_new_file.csv')
```

> **Export the `ga_schdist_raw` object to a csv file. Name the file "ga_schdist_clean.csv"**

---
# Recap

- We have now covered the basic workflow of most data projects using statistical programming software

- Launch application (RStudio) -> load any packages & import data needed for analysis -> analysis -> save/export any documents, graphs, tables, data, etc.

- Now we can focus on the skills useful for actual analysis

> **Upload your Rmd to the RLab assignment dropbox on eLC.**