Vis for Journalists Fall 2015 Lab/Assignment 2

Back to Vis for Journalists 2015 Home
Out: Sep 22 2015, 9am. Due: Sep 29 2015, 9am

Demo Resources

Other Resources

Development Aid Dataset

The use of visualization is pervasive in the media: explanatory diagrams in magazines and textbooks, graphs describing statistics and budgets, images showing spatial layouts of objects, new experimental data plotted against theoretical expectations, etc. In each case, the author of the visualization tries to convey a point of view by emphasizing some aspects of the data while toning down other aspects. The result can vary widely, from informative to misleading. How do you fare given such a task? Can you apply the design principles we talked about in class? In this assignment, you will design a visualization for a data set and provide a rigorous rationale for your design choices. The data set summarizes debt relief efforts by various countries. The money is split into administrative costs, spending on students in the donor country, spending on refugees in the donor country, spending development awareness in the donor country, and actual debt relief.

Helping third-world countries effectively is a tricky business. The article asks "So how much aid reaches developing countries? A number of groups have over the years tried to chip away at this critical, multibillion-pound question, but so far the answer has remained elusive."

Your task is to design two static (non-interactive) visualizations using Tableau that you believe effectively communicates this data with respect to two aspects that seem to be important to you.

Storytelling with Your Choice of Dataset

You will now tell a story using a dataset of your choice from the set below. Pick the data that is most interesting to you and that you believe will allow you to tell a good story. Load the data into Tableau and explore it. Create several worksheets to explore your data and keep track of the questions that you come up with. Your goal is to create several visualizations that answer at least three interesting questions. Pick your visualization types based on the kinds of questions you are answering, as discussed in the Tasks section (for example comparison, trend, distribution, correlation, and so on). Pay attention to whether your visual encodings are effective.

Available Datasets:

Writeup

For each dataset, provide a write-up describing your design and your findings.

As different visualizations can emphasize different aspects of a data set, you should document what aspects of the data you are attempting to most effectively communicate. In short, what story (or stories) are you trying to tell? Just as important, also note which aspects of the data might be obscured or down-played due to your visualization design.

In your write-ups, you should also provide a rigorous rationale for your design decisions. Document the visual encoding you used to spatially arrange the data and that choice is appropriate for the data. How did your decision facilitate effective communication? Try to apply the design principles discussed in class so far.

Also include one paragraph about the process you used to do the exploration and analysis in Tableau. Did anything limit or frustrate you? If nothing did, perhaps there was something that was more difficult than you thought it should be.

Note that understanding the data and designing the visualization may take some time, so plan accordingly.

Submit a document in PDF format, where your text writeup is illustrated by screenshot images of all of the visualization that you created. An informal style is fine, but correct grammar and spelling are required. Send by email to tmm@cs.ubc.ca by 9am Tue Sep 22, with subject "JOURN Week 2".

Credits: Part one of this assignment is based on Robert Kosara's demo and a datablog article from The Guardian. Part two is largely based on an assignment from Hanspeter Pfister's Vis course.
Tamara Munzner
Last modified: Tue Feb 16 18:45:15 PST 2016