questions

When you analyze data, there are times when a trend, pattern, or outlier jumps out and smacks you in the face. Or, you might calculate results that seem surprising. Maybe they’re real, but maybe not.

The point of visualization is to understand what data is about, which is rarely just about the numbers and almost always about what the data represents.

In a sea of endless possibilities, narrow down your visualization options by looking for what works instead of looking for what is the best.

If you just always assume the data is right, you’ll probably finish with garbage.

Answer the questions. Produce more focused and useful charts.

Be skeptical of your data in the beginning to save time in the end.

As a way to explore how people use questions in their writing, a…

For Wired, stat professor Jeffrey Rosenthal answered statistics questions from Twitter, such as…

When you visualize aspects of the data instead of just the data itself, what you show grows more obvious.

Zan Armstrong, Ian Johnson, and Mike Freeman for Observable wrote a guide on…

It still amazes me that you can give multiple people the same dataset and the results can vary depending on questions, goals, and audience.

Use basic statistical methods to move you towards fancier things.

Last month I did a short Q&A about FD and my workflow. I thought I’d elaborate on one of my answers.

On the surface, the decennial census seems straightforward. Count everyone in the country…

There’s a technical component of visualization that leans towards code, data formatting, and clicking the right buttons in the right order. Then there’s everything else that makes okay charts into something much better.

Data represents the real world, and visualization represents data. But sometimes data and the real world disagree with each other.

Make charts with a purpose.

When you don’t know where to start with a dataset, try to come…