Good visualization needs little translation

on March 31, 2010

I do not speak Portugese. I want to make that clear because otherwise, this blog post will seem a little odd. However, after seeing a Google Alert with the Portugese title "Micos de Realejo, Engenharia de Software" (Software engineers?) and Tableau Public in the text, I decided to click in. Although I could not read the text, the author provided an excellent interactive visualization which told the entire story, even for someone without the proper language skills. The visualization and my unaided analysis are below.

After skipping over the text, I read the left side of the viz to try and understand what it was depicting (besides software engineers). It read Experiencia Medio, Satisfacao Medio, Salario Medio and Numero de Respostas. Using my super-cognate skills, I guessed these were referring to Average Experience, Satisfaction, Salary and Number of Responses. Similarly, Regiao seemed to signify region (after all, they were SE, NE, S, etc). The only mystery was the bottom axis... Estado. State, of course. I guessed that this viz was about Brazil, mostly because it is the most populous Portuguese speaking country.

So, in about thirty seconds I had my bearings and could start analyzing. I found out that respondents from the CO region (central?) have very high salaries, satisfaction and experience, especially in the state DF. Seeing that I could also filter by education on the right, I selected Mestrado (masters?). Turns out software engineers with a a masters degree should avoid working in SC where they are less than half as satisfied as any other state.

"But Ross," you may say "why would I ever need to know such arcane and useless knowledge". Never! Probably. The point is this: if you have information you need to convey to a wide audience, visualization will make it through the language barrier enough to get the idea across. Skeptical? Just try reading the article. Trust me, visualization is easier.

What we like about this viz

Small multiples! People love stacked bar charts but they can be very confusing and hard to compare. These small multiples convey the message in a simple and easily comparable format.


Thanks for the kind comments. Your analysis is spot on :-).
I really loved playing with Tableau Public to create the visualization. It's truly awesome software, congratulations to the whole team. I've only scratched the surface of what can be done with it but the little I've learned has been very useful.

Congratulations again on a great product,