Lab 06: Data Importing & Cleaning

Intro Survey

data cleaning

data wrangling

data visualization

survey data

Goals

In this lab, you will…

Develop proficiency cleaning survey data
Consider implications of question types (e.g., multiple choice, select all that apply, open-ended) on data quality and analysis
Wrangle data to create desired visualizations
Discover interconnections between data cleaning, analysis, and communication strategies

Getting started

Go to our class GitHub repo and download the .qmd file for this lab.
Refer back to Lab 01 for instructions on how to get started on a lab.
You will work in your Lab 06–07 groups (see Blackboard).

Packages

We will use the tidyverse package for this assignment, and you’ll need lubridate to work with dates. If you wish to use the viridis color palettes, you will need the viridis package as well.

library(tidyverse)
library(viridis)
library(lubridate)

Data

The data come from an intro survey I give my Intro Stats students. Emails and names have been removed for anonymity. We’ll read in the data from a url:

survey <- read_csv("https://docs.google.com/spreadsheets/d/e/2PACX-1vSWWS4oNf69Rhx9COxKH445CXPl2xq7FpyabaMFpib8f9bFBFDuznskYfWIi97e8lQYl-q1M6zZ62RR/pub?output=csv")

The data are currently in the form exactly as provided from a Google Form. You will need to take many necessary cleaning steps to get it in a form that is appropriate for analysis.

Exercises

Exercise 1

Take a look at the column names.

names(survey)

Create a vector of new names using c() and save this vector as new_names. You should choose concise and informative names.

new_names <- c("variable_1", "variable_2", ...)

Then use the code names(survey) <- new_names to covert the existing column names to the new names.
Inspect your data to ensure that survey now has the desired variable names

Exercise 2

Using the glimpse() function will reveal that many of the variables are stored as characters, including the Timestamp variable.
Use the function mdy_hms() inside a mutate() function to convert the timestamp variable to a date.
Then create a histogram of this new variable.
What does this reveal about the data?

Exercise 3

Reproduce the following visualization. Note, you will need to re-order the levels of your year variable.

Hint: theme_light() and the color navyblue were used

Exercise 4

What is the distribution of gender in Intro Stats?

Create an appropriate visualization.
Make decisions about how to clean up the levels and combine as necessary. A case_when() inside a mutate may be useful.
Comment on the pros and cons of making this a multiple choice vs open-ended question in a survey.

Exercise 5

“On a scale from 1 to 10, how anxious are you about taking MATH 130 - Intro to Statistics?”

Create an appropriate visualization or summary table to explore this variable.

Exercise 6

Reproduce the following visualization. Note, you may need to re-order the levels of your variable.

Hint: Set y = 1 inside your aes() function. And add a theme layer where axis.text.y and axis.ticks.y are set equal to element_blank()

Exercise 7

What majors are represented in MATH 130?

Create a visualization to explore this variable.
Make decisions about how to clean up the levels and combine as necessary.
A case_when() inside a mutate may be useful. Comment on your decisions.

Note: ChatGPT/AI can be helpful in providing an initial solution to a tedious task like this, but make sure to review its suggestions and override if necessary - you are the analyst, responsible for analysis decisions!

Important

Select all that apply questions in a survey pose some difficult wrangling tasks. Consider the laptop variable. Suppose we want to explore what percentage of students have each of the three types: Mac laptop, iPad, or PC/Windows laptop.

The following code creates three new variables to indicate if a student has access to each type. Note you will need to adapt this code if you named your variable something other than laptop.

survey <- survey |> 
  mutate(Mac = if_else(grepl("Mac laptop", laptop), "Yes", "No"),
         iPad = if_else(grepl("iPad", laptop), "Yes", "No"),
         PC = if_else(grepl("PC/Windows laptop", laptop), 
                      "Yes", "No"))

The grepl() function searches for a given string (e.g. “Mac laptop”) in a given object (laptop). It returns the value TRUE if the string is present, and FALSE if it is not. Combined with the if_else() statement, this results in the new variable having the value “Yes” if the string is found and “No” if it is not.

Exercise 8

Use a pivot_longer to pivot the columns Mac:PC.
Consider using names_to = "laptop_type" and values_to = "laptop_access" to name the new pivoted columns.
Then, recreate the following visualization.

See the GIF below as a reminder of how pivot_longer() works

BONUS 1

Create a new variable called semester that uses the time variable to designate whether each student was in MATH 130 in a Fall or Spring semester. . Use this new variable to investigate whether students in the Fall or Spring semester were more anxious about MATH 130 on average.

Hint: the function month() applied to a date will extract just the month, year() will extract the year, etc, which may helpful in determining semester.

BONUS 2

Attempt to recreate this visualization as closely as possible.

Hint: you will first need to use a pivot_longer and then use a case_when() to recode the levels of the new variable to be the full statements seen on the y-axis).

Submission

Before submitting your .html (as a .zip file to Blackboard):

Check your code for neatness - add spaces and line breaks where appropriate to improve readability
Check visualizations for clean titles and labels
Suppress extraneous messages/warnings (e.g. set #| warning: false, #| message: false inside code chunks)
Ensure exercises are clearly labeled and your text responses are visually distinguished
Confirm neat organization and readable structure

Render one last time, check the .html file for accuracy, then convert to .zip file to upload to Blackboard.

Grading (50 pts)

Component	Points
Exercise 1	4
Exercise 2	4
Exercise 3	6
Exercise 4	6
Exercise 5	6
Exercise 6	6
Exercise 7	6
Exercise 8	6
Reflection prompts	3
Workflow & formatting	3