From the expert: 3 steps to filter your data
When it comes to visualizing data, Nathan Yau of Flowing Data has some wise words:
“Data has a lot – often too much – to say.”
What he means by this is that it can sometimes be overwhelming to know where to start with your data, ask questions of it, and eventually make meaning from it.
This post is adapted from one on Flowing Data so you should go ahead and read the original, too.
Don’t let your data ramble
An often-quoted saying in data analysis is “let the data speak for itself.”
The problem is this can lead to confusing data visualization because people think the best way to report and analyze the data is to strip everything away.
What we end up with is confusing charts at best, utter incomprehension at worst.
It’s definitely good practice to filter data, but that doesn’t mean you need to strip everything away. There is such a thing as too much data filtering for visualizing data.
You must guide the data to create value.
Nathan has a three-point process to help this value creation:
Ask the data questions
Start with the visualization basics
- Ask the data questions
The quickest way to gain meaning from data is to ask simple questions. These can lead to context-specific questions and ultimately help you create actionable insights.
- Start with the basics
Don’t jump straight in with the most complex queries or data visualizations. Start small and work your way up, asking increasingly advanced questions and creating increasingly complex queries.
Always remember the reason you are looking at this data and use that to help you focus on creating insights you can then use to make decisions. At the end of the day your data should help you understand why things are happening, so don’t get lost and forget your end goal.
Your data can say a lot so make sure your data visualization tools are helping you say what you want.