It has been estimated that an average individual receives information amounting to 17485-page newspapers every single day. Most people in the course of their working day need to access, collate, read, use, and present even more reams of information. With most professional arguments making use of large amounts of complex data, the chances of a person getting his or her point across relies on the effectiveness of presentation.
This is a skill that can be learnt. The Knight Center for Journalism offers a free 'Massive Online Course' (MOOC) that offers to teach this skill. This article presents the experiences of one such participant and also provides links to learn more about data visualisation for interested readers.
The course aims to help the participant to answer the following questions:
- How do I 'read' graphics ?
- How do I select effective means of presentation ?
- How to create interactive visuals ?
Background and motivations for participating
I work a lot with numbers in my work with the India Water Portal. In fact, a major chunk of my work is spent either searching for stories within data or more often to effectively present stories within articles/papers, that we source from various online and offline sources. Given that the manner in which people scan and process articles is changing, it makes sense to learn to present stories visually. And, data visualization is fun !
Dr. John Snow's map of deaths from a cholera outbreak in London, 1854, in relation to the locations of public water pumps is the most famous example of early data visualisation
Format of the course
The course depends a lot on interaction among the participants. Every week, we received an email with links to the readings, video lectures, and projects for the week. Separate forums were also set up and the participants are expected to discuss the lectures and projects and also critique each others work. I took the course in October-November 2012.
Extent of participation
I need to add this section because I did not participate in it fully - attempting all the exercises especially in the second half of the course. I read and watched the readings and viewing materials and also some of the forum discussions.
Week 4 onwards, the volume and nature of the tasks seemed intimidating and I focused on the basics I needed for my work with the Portal. That said, some of the participants have done good work using paper and pencils - which is very inspiring.
The course organisers suggest that approximately 6 hours a week for a total of 6 weeks, should be spent on the course; I spent maybe half that time.
What I learnt
Week 1: Here we explored what visualization is, and the difference between data visualization and infographics. To sum it up, visualization is just that -presenting all the numbers you have in a way that is understandable. An infographic is data edited to tell a story.
There was also some discussion about what a visualization needs to do, and the various forms that help one to do it. The exercise was interesting, in that we were supposed to critique a graphic. For me, the most useful part was reading the problems others saw in the visualization. It helped me understand what people expect from information, which hopefully should guide not only any charts I prepare, but also the writing I do.
Week 2: Discussions in this week were around what graphics are expected to do, and focused on the next step - how to select visualization methods. The reading is dense but useful. It presents the following checklist of questions that we need to ask before selecting a chart type:
- Does it make clear the relationship between the quantities
- Can the proportions be visually estimated
- Can we compare and rank the quantities
- Is it clear what the user is supposed to do with the information
It also describes how the human mind sees and links information - something I found useful and fascinating. The lectures illustrated the themes in this reading and explained the thought that needs to go in before finalising a visualization. This is interesting, because it is something we all do to some extent or other. The lecture inspires one to take this process a little further.
Week 3: From this week onwards, we began to talk of interactive visualization and how to plan for them. The reading seems to me to actually belong to the earlier week, because it illustrates the thought process behind an illustration. We were expected to design an interactive visualization using data on development aid.
Week 4: There were two things going on this week - more creation of interactive maps, and a reading on thematic maps. Of these, I found the reading very interesting, but also quite tough. It does not merely speak of how to use maps, but also the mathematical tools required. During week 1, we learnt where not to use maps, here we learn where to use them and how.
In addition to an overview of the types of thematic maps (and where each theme is best used), there is also a rigorous explanation of how best to present data. Are the numbers relevant ? Or a percentage ? Or a ratio ? Can you group them ? Where to split the groups ? This section is heavy on use of statistics, but useful if one hasn't learnt the science - or learnt and forgotten.
The next part looks at the mathematics behind symbols – how to size them, how to make them readable, how to present them. For me atleast, this was difficult to digest in one reading, but it helped sit down with a pencil and paper - and to keep by my side, the next time I want to plot something.
Weeks 5 and 6: These were dedicated to a final project, which I am afraid I did not do. The Dallas Morning News style guide was shared, which is an extremely detailed list of the fonts, colours, wrap styles etc to be used in various cases.
The interesting part was the extra lecture on the ethics of data visualization. This doesn't just look at deliberate misinformation, but also the cases where assumptions are presented as proofs - something I too have been guilty of. It explains the most common errors and how to avoid them - I would recommend this all who work with data.
Inspiration: David McCandless on the beauty of data visualization
Related links
