Scientific Data Presentation

PhD-level course — 2 hp — October/November 2014 Centre for Image Analysis, Polacksbacken, Uppsala

Schedule

All sessions will be held in the CBA lecture room, Polacksbacken 2115. The start times below are sharp (no academic quarter). Please come on time!

Wednesday, 8 October 2014, 9:00-16:00
Exercises to do before this session:	Find two or three difficult to understand graphs in the literature. What is the main goal of each graph?
Activities:	Lecture: Introduction: purpose of graphing, when to use tables instead, simplicity, story. (slides) Lecture: “Proper” formatting of tables. (slides) Lecture: How good are we at percieving quantities when represented in different ways? (slides) Lecture: Basic graph types: line plots, scatter plots, box/jitter/violin plots, dot plots, bar charts, pie charts, bubbles, etc. (slides) Discussion: we'll look at the difficult graphs you found, discuss them, and think of ways to make them simpler.
Wednesday, 22 October 2014, 9:00-16:00
Exercises to do before this session:	In one issue of a journal, find the three best graphs and three worst graphs. Discuss why they are good/bad. How would you improve them? Take one of the difficult graphs you found, and produce what you think is the best presentation for that data.
Activities:	Lecture: Data transformations: logarithmic axis, smoothing, aspect ratio, scale breaks, projections. (slides) Discussion: present the graphs you created, we discuss good and bad aspects, think about how to improve them. Lecture: Labels, legends, decorations. (slides) Demo: Overview of most common graphing software. (slides) Discussion: we discuss the good and bad graphs you found in the literature, and propose various ways of making them better. Demo: Using Inkscape to prepare your graphs for publication.
Wednesday, 5 November 2014, 9:00-16:00
Exercises to do before this session:	Read Edward Tufte, The Visual Display of Quantitative Information, 2nd Ed. Take some of your own data and create a journal-ready graph from it. Use Inkscape to make it prettier.
Activities:	Discussion: present the graphs you created, we discuss good and bad aspects, think about how to improve them. Lecture: Colour perception (by Gunilla Borgefors). (shortened slides) Lecture: Plotting multiple data sets and multidimensional data: small multiples, scatter plot matrix, images, surfaces, parallel coordinates, etc. (slides)

Note: Links to slides in the schedule above require a password. Students to this course have received this password by email. Please don't share the slides, they contain copyrighted material that I have collected from many sources (example graphs mostly). It is legal for me to show this material in class and make a copy for each of my students, but it would be illegal to distribute this material outside of the classroom. Thanks for your understanding!

Signing up for the course

The course is finished. The next instance, if all goes to plan, will start October 2015. Please read below and send me an email to be put on the wait list.

I applied for two instances of the course this year, but due to a misunderstanding with FUN, there will only be one instance. Next year we'll do better!

Email me at cris at cb dot uu dot se to be added to the wait list for the course. Please include the name of your department and division.

The course is limited to 20 participants. I am looking to fill this with participants from many different backgrounds, and therefore will try to limit the number of participants from the same division. In principle people will be accepted on a “first come, first served” basis, but you might be put on the wait list if too many people from your division are already accepted.

If you are registered for the course, but cannot participate for whatever reason, please let me know as soon as possible so that someone else can take your place!

Information

See here for the course description and goals.

To sign up for the course, read at the bottom of this page.

The course is limited to 20 participants, and taught in English. All sessions will be held at the Centre for Image Analysis (Polacksbacken, house 2, ground floor, room 2115).

Teachers:

Cris Luengo (responsible)
Gunilla Borgefors

Prerequisites:

No specific knowledge is required, besides that expected of any PhD student.

Literature:

Mandatory reading:

Edward Tufte, The Visual Display of Quantitative Information, 2nd Ed. (2001). [Amazon.co.uk]

Suggested reading:

Edward Tufte, Beautiful Evidence (2006). [Amazon.co.uk]
William S. Cleveland, The Elements of Graphing Data (1994). [Amazon.co.uk]
W.E. Russey, H.F. Ebel and C. Bliefert, How to Write a Successful Science Thesis (2006). [Amazon.co.uk]; chapters 20 and 21 deal with tables and figures, respectively [PDF].
Gunilla Borgefors, Ögonblick i färg – om seendets natur och naturens seende (2007), in Swedish.
T. Moeller, J. Huang and others, Introduction to Flow Visualization (slides).
booktabs manual (first three pages are good reference for proper formatting of tables)
Mastering the Art of Scientific Publication—20 Papers for 20/20 Vision About Publishing, virtual issue of the Journal of Physical Chemistry that collects relevant editorials; one editorial is on Graphical Excellence.

Software:

Besides the graphing software you are used to, you will learn to use Inkscape, an open-source drawing program, to make your graphs publication-ready. You will have to install this on your computer to do some of the homework. If you have access to a program such as Adobe Illustrator or CorelDRAW, feel free to use that instead.

Other resources:

Points of View, a series of columns in Nature Methods about graphing.
Junk Charts, a blog about terrible graphs.
Top ten worst graphs.
Color Brewer 2.0, an applet that helps you select a colour scheme for your graphs.
Vischeck, a program that simulates various forms of colour blindness.