Graphic showing distribution of songs by year and country of origin.

Mining the data from the top one hundred songs not only adds a unique visual appeal to the story, but also uncovers other underlying facts and new stories within the data. Charting the songs, it’s clear that the 1990s was the most favoured decade for songs, and that 1997 topped the list.

The two most interesting visual stories to grow from the data were charting songs across the years, and also the country of origin of the bands – a pie chart that makes the domination by UK and US artists obvious.

A few other interesting facts popped up while working with the data, but didn’t fit into a simple chart. The wide horizontal format of the years graph and the pie chart of the countries left a space on the left that worked well for a simple bullet list of numbers highlighting these random facts and rounding out the graphic.

Repeating the process

Graphic showing top artists by state, and friends per local artist.

MySpace Australia’s list of their top 20 artists by state presented a great opportunity for the same approach, thankfully this time with enough lead time before release that we could really dig into the data.

Mining the data provided in their press release and adding to it with more research to fill out details such as number of friends, record label details and location, three compelling stories came to the surface. Visually it’s clear that over half of the artists were signed to major record labels; almost all states had more international than local bands in their top twenty; and that when comparing the number of friends for local artists, there was a clear hierarchy.

A couple of other interesting views came through in the data that we didn’t use in the final chart. For example, all the Australian artists combined don’t have as many friends as just T.I, the top International artist.

Looking at the distribution of artists across the different state top 20s it’s also very clear that the majority of artists only appear in the list for one state.

An unused, alternate version of the MySpace Top 20 data.

With the other three MySpace charts being both clearer and much simpler to understand, neither of these last two visualisations made it through to the final story.

Through this process of charting and visual editing, the Hottest 100 and MySpace Top 20 charts turned otherwise text-heavy stories into richer, more engaging coverage.

Nicholas Rapp has written a post over at his personal blog discussing the creation of a graphic on American vehicle carbon dioxide emissions. The creation of this graphic demonstrates how data can be interpreted in different ways depending on the method that it is displayed.

The first map showing emissions come from major cities

In his post you can see a map of the USA with two versions of the same data overlay; the first has the total emissions per county per year, the second has total emissions per capita per year. The interesting thing is the two stories that the data tell; naturally the first says that all the emissions come from the major cities. But when you look at the second map, there as a noticeable difference in the indicators.

The second map tells a different story

Cities like Los Angeles have a large emission marker on the first version (per county) but a basically non-existent in the second version (per capita). Where as places in the middle of the country have a much larger indicator in the second version. I’m not going to go into what the data is implying (like less people in the country and further to travel etc) but it does demonstrate that from the same data two different stories can be told.

I think it is important to know what story you want to tell when designing a visualisation around data, though sometimes it may not hurt to test the data against some different filters/categories to see if there is something else that maybe more interesting.

Update

Carbon emissions from road transport

Since this post is about vechile carbon emissions, you might like to have a look at a data set The Guardian has released on carbon emmissions from vechiles around the world. From this data they have also produced a really nice clean graphic with the 10 largest polluting countries.