Data Scientists Apply Within – The Environment
As I watch the development of energy, carbon, water, and resource management tools that have flooded the marketplace recently, I can’t help but feel a little disappointed. They tend to excel at data integration with little focus on analytical tools, or they’re great at analysis and you struggle to fit your particular data set into their framework. Worst of all are the glut of offerings that are all form and no substance – flash animations where it is clear the designer has no fundamental understanding of the process that the data is meant to describe.
That’s why I was pleased to see this post from Nathan at FlowingData (be sure to pick up his RSS feed so you don’t miss another post!). He expands on the evolution of the “data scientist”, a person trained in a host of skills that take information from raw inputs to beautiful and meaningful visual display. The framework, according to Ben Fry, looks something like this:
This data-flow makes sense to me, and it’s probably easy to pick out where each of us struggles with the data sets we use. Who hasn’t wrestled with Excel, trying to fit square data into a round spreadsheet, or with a company database, attempting to extract trends from a million lines of output. On the other end, many of us try and fail to tell the story of our data in a style with visual impact, but without obscuring the real underlying meaning. The elusive “Data Scientist” hones his talents in each of these areas in order to make the numbers sing.
There’s another step in this process that a truly stellar data scientist can impact, and that is on understanding. These categories as I read them include statistics, but not a fundamental familiarity with the underlying process that the numbers represent. Sure, a data scientist can mash-up kilowatt hours and geospatial data, but does he understand them enough to find meter errors or to design an innovative visualization that tells a new story hiding behind the digits? It’s precisely this analytical understanding that seems to be so lacking in many ecometric offerings these days.
The ecometrics field will undoubtedly expand, and one hopes that this attracts a strong influx of true data scientists. In the meantime, let the buyer beware.
photo courtesy of Emilio Segre Visual Archives
Can't get enough? Try these related posts:
