What does it mean to work in data visualization?
What do data visualization professionals work on, think of and wish for nowadays?
Those questions made me looking into the data of 2017 Data visualization survey which was put together by Elijah Meeks et al to “find out what doing data visualization professionally meant”.
First I want to understand who the participants are. Data shows they are 61% North American, 29% European; and 78% males. They come from different sectors (public and private), and include both experienced professional and fresh hires.
The data covers a variety of topics such as what tools are used, how visualization is consumed, who do they see as opinion leaders, what do they want to improve in their job, etc.
Line chart and bar chart are most commonly used chart type.
Two diverging trends co-exist in dataviz: complexity and simplicity:
- Complex visuals are typically more visually-appealing and attention-grabbing, but are apparently more time-consuming. They push the boundary and can be borderline artworks. Some have more lifetime than others.
- Simple graphs remain a staple with their advantage of universal understanding. They are easy to make and quick to grasp, which also meant they can be mass produced. They typically serve the purpose of getting the idea through via one-pagers or executive presentations and are used in day-to-day settings, though they don’t always leave a lasting impression for the lack of quirkiness and uniqueness.
Analysts as well as executives are primary consumer of visualizations.
The primary consumers of the visualizations are analysts, executives, and general public, often several groups at once. As such data visualization sometimes need to serve multiple purposes or require tweaks for different audience. While visualizations are heavily consumed in business settings, there are also visualizations built for specialized audience such as medical professionals.
Most frequent form of usage is presentations.
Data visualizations are most often presented in presentations and dashboards. It’s great current state of presentation tools could take in more dynamic contents. At the same time one shouldn’t under-estimate the ‘laziness’ of the users, some of whom prefer digested statics that requires no clicking presented to them in slides. Other formats range from website to Scrollytelling which supports multimedia contents.
Data visualization are used for a multitude of purposes including analysis, communication, marketing and machine learning.
Despite common usages like analysis and marketing, I also saw some interesting use cases like teaching and entertaining (really intrigued now).
People regarded as thought leaders in data visualization bring about new tools and styles besides heavily disseminating the knowledge.
When asked “Who do you look to as a thought leader in data visualization?”, Mike Bostock, Tufte and Alberto Cairo most frequently appear in respondents’s answers, which may be attributed to the tools they built, the books they wrote and the style they created. Further down the list are many more awesome people making a dent in visualization world.
81% of data visualization professionals are self-taught.
This is at the same time expected and encouraging. Data visualization is typically infused into curriculums like data analytics/data science instead of being a field of its own, and many analysts sort of pick it up on the job. However, while the threshold seems relatively low, being able to make compelling graphs still require in-depth understanding of visual perception and narrative arc as well as great empathy with users’ needs.
The survey can be found here.
I’d like to take another deeper look at the survey results next time. The code is on github. Thanks for reading.
This is the #day7 of my #100dayprojects on data science and visual storytelling.
