A key focus of my blogging platform is technology. In today’s digital world, everything involves some form of data generation, analysis and processing. Regarding analysis, prepping the data so that it can be visualized by others is key in terms of communicating its meaning. The following contributed post is entitled, Four Tips For Data Visualization.
* * *
Computing power has largely solved many of the problems relating to collecting and processing data. Computers have not, however, solved the problem of effectively communicating the results of data analysis. That still comes down to human skill and judgment. Here are four tips to help.
Start from the raw source data (if possible)
Starting from the raw data will eliminate the possibility of you repeating any errors which have been made by other people. It will also keep your data visualization from being skewed by any unconscious bias in existing visualizations of the same data.
As a rule of thumb, this is most important with older data. Studies that were considered cutting-edge in their day might be viewed as seriously flawed now. They might also have been interpreted in the light of outdated assumptions and perceptions. It does, however, still apply to data provided by modern analytical laboratory services.
Think about your audience
Data visualization is usually easiest when you have a clearly-defined audience of people with a similar level of knowledge. It doesn’t really matter if they know nothing about the topic or if they’re experts. What matters is that they all have approximately the same level of knowledge.
If you’re catering to an audience with a broad range of knowledge, then you probably want to create multiple data visualizations. Start with one which assumes no/the least knowledge and then work up from there. Signpost these clearly so people can see which one(s) are likely to be of most benefit to them.
Focus on clarity
One of the biggest errors in data visualization is to try to fill visualizations with too much information. This may be a legacy from the days when everything was on paper so printing costs were always a concern. These days, however, data visualizations will almost always be viewed on a screen so make the most of everything this offers.
Also, be aware of its limitations. With printing, you have complete control over the use of color. With digital visualizations, by contrast, you’re dependent on the screen to reproduce the color faithfully. High-end screens generally do this very well, but lower-end ones are unlikely to be so accurate.
This means that you should usually avoid relying on conveying fine details by subtle changes of color. Instead, you generally want to use as much contrast as possible. This also goes for the text and the background. Furthermore, all text should be in simple fonts. It’s fine to bold important words occasionally, but avoid italics and regular use of capital letters.
Look for ways to make visualizations interactive
This is only possible if a person is going to view a data visualization on their own device. This is, however, likely to be the case a lot of the time, so it’s always worth considering. Interactive data visualizations follow the same principle as having a person click on a link for more information. Instead of going to a new page, however, a part of the visualization expands.