Design Exercise 2: Make some visualizations

For this second exercise, we will build on the first to actually try to make “good” visualizations. You need to pick one data set (from a list) and make two visualizations from it. We’ll ask you some questions about those visualizations to help you think about them. This is a warm up to doing it with a bigger and harder data set for the Design Challenge.

We want you to use Tableau for this assignment. You should have experience from the previous exercise. But, a reminder about Tableau access and resources is on the Tableau page.

What you must do:

  1. Pick one data set from the list below. (this is the same list as Design Exercise 1: Try Tableau). We will provide data files.
  2. Make two visualizations from this data set. They need to be different.
  3. Answer two questions about each visualization (described below).
  4. Upload your answers and visualizations to the Canvas Quiz: Design Exercise 2: Use Tableau (due Mon, Oct 4).

The “quiz” will ask you two questions about each visualization. You should consider these questions before you make the visualizations (or at least as you make the visualizations).

  1. What does this visualization make easy to see? I like to think of this as the “story” that the picture tells. I was going to phrase this as “what were you trying to make easy to see” - but I am open to the idea that your answer might be post-hoc: you made something and then by self-critique realized that it was good at something.

  2. What design decisions did you make that caused this (the thing that you say is easy to see)? This gets at the reasons why #1 happened. What did you do?

As a practical point: write your answers off-line and copy them into the canvas quiz (since Canvas isn’t good about letting you edit after you submit).

The Data Sets

This list is the same as the list for Design Exercise 1: Try Tableau. You may use one of the data sets that you used for that previous assignment.

For DE1, you need to pick 2. For DE2 you need to pick one (it can be one of the ones you picked for DE1).

To make things more convenient for you, we will provide all of these as Tableau Online DataSets - so you don't need to upload them yourself. If you aren't using Tableau online, you will need to get your own copy of the data set. We'll post instructions on how to access our uploaded versions somewhere else.

The data files (as CSV) are all in a Canvas Files folder

  1. Pizza (Barstool)

    • documentation: <https://github.com/rfordatascience/tidytuesday/tree/master/data/2019/2019-10-01>
      • this has other data sets as well
    • <https://github.com/rfordatascience/tidytuesday/raw/master/data/2019/2019-10-01/pizza_barstool.csv>
    • 464 lines
    • might not be enough to tell stories, but good enough to experiment with
  2. IMDB 5000 Moview (Kaggle Data Set)

    • originally from Kaggle
    • this was an option for the overall challenge in previous years
    • there are 5000 movies (rows), and a good number of attributes of each movie
  3. NYC Air BNB Data

    • originally from Kaggle
    • 49K rows, 7MB
    • this is big enough to be interesting, but the set of features may be too limited to tell interesting stories
  4. 2017 President's Budget (last year we have the data easily available)

    • these datasets have historical trends (nice to looking at multiple series) and part/whole relationships
    • lots of documentation: <https://github.com/WhiteHouse/budgetdata/tree/2017>
    • Receipts (small - 245 rows) GitHub
    • Budget Authorization GitHub
    • Outlays GitHub
  5. Spotify Music

    • 12 continuous features, a few catergorical features, plus names
    • 32K songs, 8MB CSV file
    • features well documented in GitHub ReadMe
    • GitHub Repo w/README
    • CSV File
    • This dataset is big enough to be interesting… but it isn't clear that good stories will emerge.
  6. Students Choice

    • I wasn't going to allow students to come up with their own data sets. But, for this exercise your can. You may only do this for one of your two DE1 data sets. And there are rules.
    • The rules you must follow:
      1. The dataset must be publicly available. You must be able to give a URL where someone can find it publicly. Other students must be able to use it.
      2. You must announce it (so that other people can use it as well). Make a posting on Piazza giving the URL where someone can find it on the web, and a brief description.
      3. Getting it into Tableau is up to you. Generally, this is easy (Tableau is pretty smart about uploading data), but you might need to do some "cleaning".
    • This is only for DE1 - not DE2.
    • You may pick a data set that someone else has already posted on Piazza (you don't need to post again).