Topics covered in this lab:

• Exploratory Data Analysis (EDA) - data visualizations and numerical summaries
• Statistical inference about a mean
• Writing a lab report using R Markdown
• Tracking changes and using git and GitHub

Clone the repo & start new RStudio project

• Go to the sta210-sp20 organization on GitHub (http://www.github.com/sta210-sp20). Click on the repo with the prefix lab-01-review-r-. It contains the starter documents you need to complete the lab.

• Click on the green Clone or download button, select Use HTTPS (this might already be selected by default, and if it is, you’ll see the text Clone with HTTPS as in the image below). Click on the clipboard icon to copy the repo URL.

• Go to https://vm-manage.oit.duke.edu/containers and login with your Duke NetId and Password.

• Click to log into the Docker container STA 210 - Regression Analysis. You should now see the RStudio environment.

• Go to File ➡️ New Project ➡️ Version Control ➡️ Git.

• Copy and paste the URL of your assignment repo into the dialog box Repository URL. You can leave Project Directory Name empty. It will default to the name of the GitHub repo.

• Click Create Project, and the files from your GitHub repo will be displayed the Files pane in RStudio.

Configure git

There is one more piece of housekeeping we need to take care of before we get started. Specifically, we need to configure your git so that RStudio can communicate with GitHub. This requires two pieces of information: your name and email address.

To do so, you will use the use_git_config() function from the usethis package.

Type the following lines of code in the console in RStudio filling in your name and email address.

⊕The email address is the one tied to your GitHub account.

library(usethis)
use_git_config(user.name = "GitHub username", user.email="your email")

For example, mine would be

library(usethis)
use_git_config(user.name="matackett", user.email="maria.tackett@duke.edu")

If you get the error message

Error in library(usethis) : there is no package called ‘usethis’

then you need to install the usethis package. Run the following code in the console to install the package. Then, rerun the use_git_config function with your GitHub username and email address associated with your GitHub account.

install.package("usethis")

Once you run the configuration code, your values for user.name and user.email will display in the console. If your user.name and user.email are correct, you’re good to go! Otherwise, run the code again with the necessary changes.

RStudio & R Markdown

Below are the general components of an RStudio project.

Below are the general components of an RMarkdown file.

To get more details about the RStudio project, RMarkdown file, and just R in general, read Getting Started in Data Visualization by Kieran Healy.

Packages

We will use the following package in today’s lab.

library(tidyverse)

If you need to install the package, run the code below in the console.

install.packages("tidyverse") 

YAML

⊕The top portion of your R Markdown file (between the three dashed lines) is called YAML. It stands for “YAML Ain’t Markup Language”. It is a human friendly data serialization standard for all programming languages. All you need to know is that this area is called the YAML (we will refer to it as such) and that it contains meta information about your document.

Before we introduce the data, let’s update the name and date in the YAML.

Open the R Markdown (Rmd) file in your project, input your name for the author name and today’s date for the date, and knit the document.

Committing changes:

Next, go to the Git pane in RStudio.

If you have made changes to your Rmd file, you should see it listed here. Click on it to select it in this list and then click on Diff. This shows you the difference between the last committed state of the document and its current state with your changes. If you’re happy with these changes, write “Update author name” in the Commit message box and click Commit.

You don’t have to commit after every change, this would get quite cumbersome. You should consider committing states that are meaningful to you for inspection, comparison, or restoration. In the first few assignments, we will provide some guidance on when to commit and an example commit message.

Pushing changes:

Now that you have made an update and committed this change, it’s time to push these changes to the web! Or more specifically, to your repo on GitHub. Why? So that others can see your changes. And by others, we mean the course teaching team. (Your repos in this course are private and can only be seen by you and the teaching team.)

In order to push your changes to GitHub, click Push. This will prompt a dialogue box where you first need to enter your user name, and then your password. This might feel cumbersome. Bear with me… We will teach you how to save your password so you don’t have to enter it every time. But for this one assignment you’ll have to manually enter each time you push in order to gain some experience with it.

Make sure you push all the files from the Git pane to your assignment repo on GitHub. The Git pane should be empty after you push. If it’s not, click the box next to the remaining files, write an informative commit message, and push.

Exploratory Data Analysis

1. Now that we’ve had a quick view of the dataset, let’s get more details about its structure. Sometimes viewing a summary of the data structure is more useful than viewing the raw data, especially if the dataset has a large number of observations and/or rows. Run the code below to use the glimpse function to see a summary of the trails dataset.

   How many observations are in the trails dataset? How many variables?

glimpse(trails)

2. Before conducting statistical inference (or eventually fitting regression models), we need do some exploratory data analysis (EDA). Much of EDA consists of visualizing the data but it also includes calculating summary statistics for the variables in our dataset. Let’s begin by examining the distribution of status with a data visualization and summary statistics.

   • What is a type of graph that’s appropriate to visualize the distribution of status? Fill in the ggplot code below to plot the distribution of status. Include informative axis labels and title on the graph.

   • Then, calculate the proportion of observations in each category of status by completing the code below.

ggplot(data = trails, aes(x = status)) + 
  __________ + 
  labs(x = "_________", 
       y = "_________", 
       title = "__________")

trails %>%
  count(_____) %>%
  mutate(proportion = n / sum(n))

3. Since we want to analyze characteristics for outdoor trails in Durham, we will just use data from currently existing trails for the remainder of the analysis. Complete the code below to use the filter function to create a subset consisting only of trails that currently exist and have a value reported for length. Assign the subset the name current_trails. (Hint: There should be 1054 observations in current_trails.)

__________ <- trails %>%
  filter(status == "______", !is.na(______))

⊕Check your lab-01 repo on the GitHub website (you may need to refresh the page in your browser) to ensure that all of the files are up-to-date.

This is a good place to knit, commit, and push changes to your remote lab-01 repo on GitHub. Be sure to write an informative commit message (e.g. “Completed exercises 1 - 3”), and push every file to GitHub by clicking the checkbox next to each file in the Git pane. After you push the changes, the Git pane in RStudio should be empty."

Use current_trails for Exercises 4 - 7.

4. Let’s examine the distribution of length. One important part of EDA is creating data visualizations to see the shape, center, spread, and outliers in a distribution. Data visualizations are also useful for examining the relationship between multiple variables. There are a lot of ways to make data visualizations in R; we will use the functions available in the ggplot2 package.

   Make a graph to visualize the distribution of length. Include an informative title and axis labels.

⊕See Section 7.3.1 “Visualizing Distributions” or the ggplot2 reference page for details and example code.

5. Next, fill in the code below to use the summarise function to calculate various summary statistics for the variable length. You can use the summarise reference page for more information about the function and example code.

current_trails %>%
  summarise(min = __________, 
            q1 = __________, 
            median = __________, 
            q3 = __________, 
            max = __________, 
            iqr = __________, 
            mean = __________, 
            std_dev = __________
            )

6. Describe the distribution of length. Your description should include comments about the shape, center, spread, and any potential outliers. Use the graph from Exercise 4 and relevant summary statistics from Exercise 5 in your description.

7. We want to limit the analysis to trails that are more likely intended for day hikes, rather than multi-day hikes and camping. Therefore, let’s remove the extreme outliers from the data for this analysis and only consider those trails that are 15 miles or shorter.

   Filter the dataset to remove the extreme outliers. Be sure to save the updated dataset, so you can use it for the remainder of the lab.

This is a good place to knit, commit, and push changes to your remote lab-01 repo on GitHub. Be sure to write informative commit message (e.g. “Completed exercises 4 - 7”), and push every file to GitHub by clicking the checkbox next to each file in the Git pane. After you push the changes, the Git pane in RStudio should be empty."

8. Consider the distribution of dogwalking.
   • What are the values of dogwalking in the dataset? Show the code and output to support your answer.
   • What do you think is the most likely reason for the missing observations of dogwalking? In other words, what does a missing value of dogwalking indicate?
9. Complete the code below to impute (i.e. fill in) the missing values of dogwalking with the appropriate value. Then, display the distribution of dogwalking to check that the missing values were correctly imputed.

__________ <- __________ %>%
  mutate(dogwalking = if_else(is.na(dogwalking),______,dogwalking))

10. Now that we’ve completed the univariate EDA (i.e. examining one variable at a time), let’s examine the relationship between the length of the trail and whether dog walking is allowed. Make a graph to visualize the relationship between length and dogwalking and calculate the appropriate summary statistics. Include informative axis labels and title on your graph.

11. Describe the relationship between length and dogwalking. In other words, describe how the distribution of length compares between trails that allow dog walking versus those that don’t. Include information from the graph and summary statistics from the previous exercise in your response.

This is a good place to knit, commit, and push changes to your remote lab-01 repo on GitHub. Be sure to write informative commit message (e.g. “Completed exercises 8 - 11”), and push every file to GitHub by clicking the checkbox next to each file in the Git pane. After you push the changes, the Git pane in RStudio should be empty."

Statistical Inference

We’d like to use the data from the trails in Durham to make more general conclusions about trails in urban areas in the southeastern United States. We will reasonably consider the trails in Durham representative of the trails in other urban areas in the southeastern United States.

Over the next few questions, will use statistical inference to assess whether there is a difference in the mean length of trails that allow dog walking versus those that don’t.

12. The following conditions must be met when we conduct statistical inference on the difference in means between two groups. For each condition, specify whether it is met and a brief explanation of your reasoning.

    • Independence
    • Sample Size
    • Independent Groups

13. While we have observed a difference in the mean length in trails for those that allow dog-walking versus those that don’t, let’s assess if there is enough evidence to consider the difference “statistically significant” or if it appears to be due to random chance.

    The null and alternative hypotheses are written in statistical notation below. State the hypotheses in words in the context of this analysis.

\[H_0: \mu_1 - \mu_2 = 0 \\ H_a: \mu_1 - \mu_2 \neq 0\]

⊕Type ?t.test in the console to learn more about the function and its options.

14. Fill in the code below to use the t.test function to calculate the test statistic and p-value. Replace response with the variable we’re interested in drawing conclusions about and group_var with the variable used to define the two groups.

t.test(response ~ group_var, data = ________, 
       alternative = "___________", 
       conf.level = 0.99) #less, greater, or two.sided

15. Use the output from the previous exercise to answer the following:
    • Write the definition of the test statistic in the context of this analysis.
    • Write the definition of the p-value in the context of this analysis.
    • State your conclusion in the context of this analysis. Use a significance level of \(\alpha = 0.01\).
16. Notice the confidence interval for the difference in mean trail length printed in the output from Exercise 14. Interpret this confidence interval in the context of this analysis.

You’re done and ready to submit your work! Knit, commit, and push all remaining changes. You can use the commit message “Done with Lab 1!”, and make sure you have pushed all the files to GitHub (your Git pane in RStudio should be empty) and thatall documents are updated in your repo on GitHub. Then submit the assignment on Gradescope following the instructions below.