Uncategorized

Visualization vs Data Configuration

Posted

Visualization vs Data Configuration

What’s Behind the Scenes?


Australia, represented as 100 people. Simpler to understand than “Australia represented in percentage points” but pretty much the same thing. Aside from the obvious issue of losing value as the grain is reduced from a detail level of 24.16 million to a mere 100, there is also the question of natural biases as to what was aggregated, and when visualizing the results in a chronological order, some obvious biases in what was presented.

In this case, there is some, but not much, popular demand driven bias in the aggregations available, this is data from the Australian Statistics Bureau, which is a function of government, and has a budget too limited to just “Count all the stuff out there”. In culling some aggregations or adding some survey variables, there is always a percentage of bias, and in including popular expectations, there is more bias.

Of course, journalists sell news stories, so they are more prone to bias in reporting what the public wants to read about. More about gender biased in the corporate sector, less about food diversity in restaurants. More about age of home ownership, less about types of business growing from small to medium. More about immigrants, less about emigrants. More about school attendance, less about school curriculums.

All the data is actually there – and much more, but while the following visual is fun to play with, try to remember the 10’s of thousands of man hours, people and metrics that were collected to get this data, the fact that there is so much more data available and the complexity in preparing the data from consumption. Then when you are finished having fun with this fast food of data visualisations, hop over to the Australian Bureau of Statistics website and see the real thing.

Getting the back end right is where the real work is, most visualization tools offer “fast food” demonstrations, high in sugar, low in nutrients. Talk to people who know how to prepare the data so that you get real ongoing deep value. Spend some time with us so we can help you bring health back to your knowledge system.