Day 16 In-Class: Introduction to Data Visualization

Day 16 In-Class: Introduction to Data Visualization#

✅ Put your name here
#

Learning Goals:#

By the end of this assignment, you should be able to:

Compare and evaluate plots

0. Introduction#

The goal of this assignment is for you to collect feedback from your group on the visualization you made for the PCA and improve it during class. You plot should be both descriptive and easy tp understand. And, ideally, your plot should tell a story. This is meant to be a test run of the skills you’ll be expected to demonstrate for the semester project.

In the pre-class assignment, you were tasked with making a plot relevant to your semester project, based on one of the datasets we’ve used in class, or from any other dataset you have handy. In today’s assignment, you will be:

Sharing your plot with your group
Getting feedback from your group about the plots
Improving the plot according to the rubric listed in Part 3 and the feedback you received.

2. Creating an effective plot#

✅ TASK Now your goal is to take the feedback you received the details below to make an informative and easy-to-understand plot.#

You plot should aim to meet the following criteria:#

Does the plot tell a story?
No wasted space
Labels, legend, and title are informative without cluttering
If possible, are you presenting multi-variable data?

1. Does the plot tell a story?#

Your plot should be able to stand on its own and inform/convince people of your point.

2. No wasted space#

You only have a finite amount of space when making a figure and presenting it, so you want to make sure that every square inch of your plot conveys information. Let’s look at an example, the plot below show the amount of cocoa beans export for each country

The problem that we run into here is a considerable amount of wasted space. Most countries export a substantially smaller amount than the countries along the west coast of Africa. Our plot is massive, but we’re getting information from ~20%, which is not ideal.

One possible way of dealing with this issue is to change the y-scale to log, which allows you to see information about all of the other countries.

The risk here is that your audience might not be as comfortable with reading/interpreting log plots. It’s functionally useless if no one can understand your plot. You will need to decide what works best for your plot.

3. Labels, legend, and title are informative without cluttering#

The x/y labels, legend, and title are the tools you will use to give your audience context, making them crucial for a well-made plot.

Let’s again look at the Cocoa bean export plot from above. Our title is a good descriptive summary that isn’t taking up too much space. We run into trouble when we try to add too much description, and we end up taking up a lot of space, potentially in attempting to be a little too on the nose with the story of our plot.

Aside from being aesthetically displeasing, too much space is being taken up by text. You want your text to balance giving enough information without becoming the center of attention. We want the eyes to be drawn to the plot first and then look at the text to get context; in this case, it’s not clear where we should focus our eyes. Also, we could probably make some improvements to how we’ve formatted the axis labels for the x-axis, right now it’s pretty hard to read those.

How might we improve the readability while also perhaps using the space more effectively?

4. Presenting multi-variable data#

NOTE: You do not need to have multi-variable data. There are just more design choices possible for this kind of plot

It will often be the case that you’ll be dealing with data with multiple dimensions that could be important. In this case, it’s important to find a way to include all of the different variables in your plot so that the reader can see all factors in play.

Let’s take a look at the figure from the pre-class assignment.

In this figure, we’re currently displaying four variables: MPG, Horsepower, Country of Origin, and Weight. This is an excellent example of a plot that has little to no wasted space, has informative text that isn’t drawing attention away from the figure, and shows all relevant variables. The figure above is the kind of plot you should be aiming for, but make sure you don’t get too carried away with trying to fit too much information into one plot – that can lead to an overload of information.

3. (Time permitting) Present your updated plot to your group#

✅ TASK If you have time, check back in with your group and seee what they think of your updated plot.#

Congratulations, you’re done!#

Submit this assignment by uploading your notebook to the course Desire2Learn web page. Go to the “In-Class Assignments” folder, find the appropriate submission link, and upload it there. Make sure your name is on it.

See you next class!