Putting Data Visualization to Work for Your Analytics Projects
Every second, 40000 search queries are processed by Google, 6000 tweets are posted on Twitter, and hundreds of products are sold on Amazon (source: internetlifestats.com). These numbers continue to increase and more and more data is created every day. Data is being generated faster than it can be analyzed. And the ability to make sense of large amounts of data is in high demand.
Why data visualization?
Glance over the data visualization below.
Even if you didn’t read the descriptions, you probably noticed the upward trend. This chart shows rising global temperatures by taking the average temperatures between 1961 and 2010 as the baseline, and comparing the temperatures between 1850 and 2016 to this baseline. You can quickly make sense of this data. Yet, there are over 100 points of information on this chart. And that’s the power of data visualization. It allows us to make sense of a large amount of information quickly.
Data visualization in the visual analytics process
Contrary to popular belief, data visualization is not simply the last step of an analysis – creating a quick chart to put on a slide for presentation to management. Data visualization is part of visual analytics.
Take a look at the chart below.
There is, even more, information on this one. It shows dietary trends for over 150 countries over 50 years, for four categories: calories, protein, fat, and food weight.
Here, we take the world average as our baseline and compare each country to this average. Imagine that you were tasked with comparative analysis of dietary trends across 150 countries. Looking at all this data in the form of a table would take a long time.
Analyzing this data using visual analytics techniques makes this process much more interesting and less time-consuming. Let’s filter to category ‘fat’ for example.
It is easy to spot which countries have something interesting going on. Look at Canada: the consumption of fat is declining. Now look at Kuwait: here, the consumption of fat is increasing. It’s probably worth diving in and looking at what’s driving these trends.
And that’s the power of visual analytics. It allows us to use data visualization not as the final step of the analysis, when we want to share our findings with the team, but as the means to making sense of large amounts of information, quickly.
Why is data visualization so powerful?
A large portion of our brain is dedicated to visual processing.
Our brains can process 10 billion bits per second of visual information. With so much processing power readily available, as soon as we see a chart, our brains start making sense of it.
The large size of our visual cortex also allows us to take care of visual information in System 1, which is a fast, active thought process that doesn’t require much focus, preparation, or effort (Source: Daniel Kahneman, Thinking Fast and Slow). Being part of System 1, data visualization is more likely to result in actions and decisions than any other type of analysis.
Where does data visualization come in, in terms of data science skills?
Data visualization is part of communication.
As a data scientist or a data analyst, you need to engage your management with clear insights and communicate effectively.
In fact, the ability to communicate is what defines a successful analyst and data scientist in the eyes of management. Knowledge of math and statistics, experience in coding, and even domain knowledge is not going to be sufficient to move the needle in any organization. Data scientists and analysts need to communicate their findings in a clear, concise and easy-to-digest manner. And as you’ve probably realized by now, data visualization does this best.
What skills will you need to get your foot in the door and start your data visualization career?
A quick search on LinkedIn for “data visualization” returned 9000 jobs worldwide in the third week of September 2017. That’s a lot of jobs! I used Python to quickly collect the job descriptions. Here is what I found.
Data visualization overlaps with other roles in data, including software development, data engineering and data science. Knowledge of Tableau, a data visualization tool, is in high demand, but the use of other data visualization tools is often required as well, including Microsoft PowerBI, Qlik, MicroStrategy, and Adobe Cloud. Ability to wrangle the data using SQL, and understanding of core databases such as SQL Server and Oracle can also be essential.
The rapid growth of data volumes won’t be slowing anytime soon. Data visualization removes barriers to data analysis. Those who can manipulate these tools will turn data visualization to their advantage and move ahead in their careers.
And of course, a data visualization portfolio will help you to get your next data visualization job. Check out open source data repositories such as data.world and kaggle.com and create something of your own. Prefer to work on a real project? Join Data for a Cause, where volunteers create data visualizations for NGOs, nonprofits and charities. You will work with real data while building your portfolio and putting your skills to work for a good cause.
Are you ready to harness the power of data visualization? Check out BrainStation’s Data Analytics course to get core skills and learn about data visualization.