Interactive Visualization Graphics for Scientists

In recent years, the interactive data visualization/graphic space has experienced several exciting developments. With these changes, it has never been easier for experts in science and data to create visualizations, rather than depending on visualization technology or design expertise to communicate ideas.

It has always been challenging to translate and synthesize concepts into graphics that succinctly communicate through an intermediary. But now scientists and researchers have the tools to create high-impact, aesthetically pleasing, and powerful interactive data visualizations on their own, or within their own teams, rather than being forced to look “outside” for help—or rely on words alone.

Data visualizations have many applications. Here are just a few:

• Communicating ideas for public consumption and understanding
• Distilling complex ideas into more “digestible” ideas for the media
• Deepening the understanding of potential funders for research grants
• Summarizing the impacts of many complex factors

In this whitepaper, we will share how scientists can leverage readily available technologies to create powerful interactive graphics and visualizations today, with limited-to-no programming ability.

In general, visualization technologies come in three categories:

• “No-code” platforms that require only data entry and no software development
• Visualization “frameworks” that allow for more complex and flexible visualizations using pre-defined components
• Fully custom code (the traditional approach)

We will cover the first two points above, and we will discuss how scientists and researchers can utilize technologies in these categories to create data visualizations, with very little investment of time to learn the ins and outs of the technologies.

“No-code” platforms

The easiest way to get started with visualizations is through no-code platforms. The closest analogue to “no-code” platforms is Microsoft Excel, which allows you to create a variety of charts and graphs from a selected range of data cells.

The main difference with no-code platforms, however, is their interactivity, complexity, and capability for connections to other web-based platforms.

Here are just a few options (but there are many others):

• flourish.studio
• datawrapper.de
• datamatic.io

A positive of this approach is that you can drop in your data and pick a display/visualization option (along with possibly one or more interactivity options, such as data segmentation options).

To share or use your results, you can generally:

• Share the resultant visualization via a URL provided by the platform
• Download an image file (png, jpg, etc.) for use in documents such as reports
• Embed the visualization in a page on your own website by pasting a provided snippet of HTML

A downside of this approach, however, is that for-profit companies develop and make these platforms available, and their usage usually entails a monthly subscription fee. Even modest sums can prove challenging—if not for the costs directly, then for the bureaucracy involved in getting the expense authorized by a given scientist’s organization.

Visualization frameworks/toolkits

The next category of visualization options includes freely available visualization “frameworks” or “toolkits,” often accessible under an Open Source license (meaning they are generally free to use, though with some limitations depending on the particular Open Source license referenced in the code base).

Here are a few options (but as with the “no-code” options, there are many more):

• d3js.org
• plotly.com (note that Plotly mostly covers their paid offering; refer to their Github page for the Open Source option)
• Highcharts.com (similar to Plotly, check their Github profile for the Open Source option)

Open Source code allows you to see and modify/customize the code. But it often requires some ability to understand and manipulate the code for your purposes.

While the “coding” aspect of these platforms might seem intimidating, especially for a non-coder, it is incredibly easy to achieve your goals through simply copying and pasting code examples from the documentation or the myriad blog posts from people just like you who have successfully done the same, or through YouTube explanation videos.

The general process for these toolkits is as follows:

• Choose a chart option from the menu of choices available
• Copy the HTML provided in the documentation
• Paste the HTML into your own webpage (note that you will need permission to paste code/HTML on your website, which may require a few extra clicks or special permission from your webmaster)
• Swap out the data in the sample code with your own data, respecting the required formatting of the previous data
• Include a reference to a special JavaScript and/or CSS file (covered by the documentation) so that your webpage knows how to render the graphic when the page loads

The above process may seem intimidating, but with experience, you can take those steps in mere minutes. A YouTube search for “How to embed a Highcharts graphic in my webpage” will yield countless step-by-step how-tos and should allay any concerns that this is beyond your abilities, no matter how meager.

Advantages of frameworks/toolkits

The main advantage of using frameworks/toolkits is that they are often completely free. The business model for these companies is to provide the code for free to DIYers and to charge only those who want a hosted platform for its ease of use.

But another advantage is that you can extend or customize your visualizations (for example, allowing your website visitor to manipulate or interact with the data in advanced ways, or for the visualization to auto-populate with data from external data stores) by modifying the code. While this will likely be beyond the abilities of the intended reader of this whitepaper, a software engineer could easily accomplish this task.

In conclusion

Scientists and data analysts have many skill sets that are complementary to creating rich and dynamic data visualizations. By exploring the tools and ideas (as well as the associated documentation for each) shared in this whitepaper, motivated individuals can create and use visualizations in their own work with no help from a software developer.

The best way to get started is to pick a platform and dive in. You’d be surprised by what you can create by experimenting with the visualization technologies mentioned in this post!

‍