Keeping Up With . . . Data Visualization

This edition of Keeping Up With… was written by Brandi Jagars.

Brandi Jagars is a Visiting Research & Instruction Librarian at the University of South Florida Libraries, email: bjagars@usf.edu

Introduction

Data visualization has become a widely utilized skill in many industries. It is a great way to demonstrate trends, show value of services, and provide information to stakeholders. Even if you are not a “data person,” understanding basic design skills that tell a story through data is a skill valuable for many academic professionals. This article will provide examples for how library data can be visualized, why it should be, and how you can get started.

Valuable Data is Everywhere!

To begin a discussion on data visualization, it is helpful to consider the different types of data an academic library may collect for their records and services. Your collections department might have spreadsheets containing titles and prices from publisher purchases. Research consultation request forms can document who is asking for what type of assistance. If your library uses a badge system for instructional courses, who takes the courses and how many badges are administered is another dataset that might be available to you. The publisher platform your library uses to display eBooks might have data on click-through and page views; data like this can help departments decide if funds should be used to re-subscribe to materials. You might even find use for data not directly collected by your library. Journal rankings and faculty publication statistics can assist with determining what journals to cut due to budget and can also provide insight on the journal’s reputability.

Data You Can See

Now that we have discussed potential data sources, here are a couple of instances in which data visualization can be utilized in an academic library:

Figure 1: USF Libraries Tampa Campus Research Consultations – Fall 2021

Figure 1 is a snapshot of a draft dashboard that I have created with Microsoft’s Power BI program. It contains data collected from the University of South Florida’s librarian’s consultation requests to show various trends within the data. Particularly important are the visuals related to patron type and frequency.

This dashboard shows that a majority of patrons requesting consults were graduate students. Noticing this trend in the visualization might be something we want to focus on when developing instructional tools. For example, graduate students often seek assistance with literature or systematic reviews, so being knowledgeable and up to date regarding library resources to assist with these projects is a good idea. This dashboard can also be filtered by individual librarians to determine which departments or colleges are requesting assistance most often. In areas where there is higher demand for consults, a faculty librarian might use this visualization to suggest their job duties be altered to account for more time being spent on research consultations.

Figure 2: By the Numbers: Textbook Affordability Infographic

Figure 2 is an infographic I created with Canva that provides an overview of student’s attitudes and actions related to the rising cost of textbooks. Historical information about the Consumer Price Index (CPI) was sourced from the Bureau of Labor Statistics.[1] Student responses to a PIRG survey regarding textbook costs were also included to show how students are reportedly handling the financial burden.[2] While this data comes from outside sources, it is utilized in such a way to communicate to stakeholders and donors the importance of the USF Libraries’ Textbook Affordability Project (TAP).[3] This visualization could be used to appeal to university leadership to continue or acquire funding for such a program. It might also be shared to advocate for a higher budget allocation for electronic textbooks which students would be able to use at zero-cost, thereby avoiding purchasing costly materials. Visualizing data with an emotional appeal can be effective as a call to action in a variety of scenarios.

The examples provided in fig. 1 and fig. 2 are just a few of the myriad of stories the data sources chosen could tell. This is important to remember when considering the who, what, and why of your story and also why data visualizations are so powerful. If you share with your audience only the raw data, what might they infer or focus on? Will they be able to see any trends or even understand departmental nuances? Clear and visually appealing charts and graphs are a powerful tool to share your message.

Getting Started

As illustrated above, data visualizations are virtually limitless in their design and purpose. This revelation can be both exciting and intimidating as you begin your project. Provided below are some general tips to consider when creating data visualizations. Please keep in mind that this list is non-linear in nature and you may find other methods personally useful once you begin your data visualization process.

Evaluate the Data: Was the data collected in a reliable manner? If from an outside source, does the author have appropriate qualifications to collect this data, and will they permit your use of the data? How often does the data update – is your version timely? When in doubt, it could be helpful to utilize elements of the TRAAP test as you might with other academic materials.[4]

Establish Questions: While you may notice other interesting trends once you begin working with a dataset, having an idea of questions you want to answer will help guide how you clean (organize) your data and what types of visuals are appropriate. Examples could include, “What are the busiest times for graduate study room reservations?” or, “How much money can a student save by using the online textbooks the library provides access to?”

Consider Your Audience: Who is the intended recipient of this information? Colleagues familiar with the data may need less explanation of its collection and purpose. Since I intended fig. 2 to be viewed by a broad audience, I decided to include how CPI is calculated to make the accompanying graph easier to understand, thereby increasing its impact.

Select the Right Tools for the Job: There are numerous programs and websites that can be used to create data visualizations. Your choice may be affected by price, availability at your university, features needed, ease of use, etc. For example, I chose Power BI for my consultations dashboard because I had the data organized into one spreadsheet that I can refresh and replace when needed. Canva is simpler to learn and offers more creative freedom but does not have the same interactivity with graphs and charts and most data has to be manually entered.

Use Visualizations Appropriately: Not all graphs and charts can effectively display the same types of data. Additionally, graphs can be intentionally (and sometimes unintentionally!) manipulated to be misleading. Venngage provides a brief overview of how altering graphs can skew an intended message.[5]

Consider Aesthetics: Do not worry— this does not mean you need to be a graphic designer! Selecting the most useful type of graph, an appropriate color palette, and easy to read fonts will ensure your visuals are appealing and accessible to all viewers. A good place to start is your university’s marketing/branding site. They generally contain specific color palettes and font styles used for university communication.

Resources

While this list is by no means comprehensive, the resources included can be helpful to anyone who wishes to refine their data visualization skills. It really is easier than you might think!

• Free Data Visualization Tools – This list created by PCMAG offers several free options for data visualization, many of which are beginner-friendly.
• Advanced Data Visualization Tools – Also curated by PCMAG, these programs are powerful and offer countless features that may be beyond the scope of a beginner’s needs. However, there are courses available online that provide excellent guidance in mastering these programs.
• Canva – While Canva is not marketed as a data visualization tool, I find it incredibly helpful for layout inspiration and infographic creation for smaller datasets.
• LinkedIn Learning – There are numerous courses available for a variety of data visualization programs. A majority of the programs listed above have multiple courses available. Before purchasing a subscription, check with your university to see if you have institutional access.
• Coolors - This generator is a great tool for creating a color palette that is appealing for your visuals. There is an option to simulate colorblindness to see if your palette is accessible.
• Information is Beautiful - If you are looking for inspiration or want to step out of the norm in your visualizations, this site is a great place to view aesthetically pleasing graphs and charts.

Conclusion

Data can be powerful, but it must also be digestible. Having the ability to visually articulate what your data means will enhance your research capabilities, reveal opportunities for growth, and communicate to stakeholders and donors the importance of the tireless effort your library contributes to your university’s mission.

Notes

[1] “Consumer Price Index (CPI) Databases,” Bureau of Labor Statistics, Accessed January 3, 2022, https://www.bls.gov/cpi/data.htm.

[2] U.S. PIRG Education Fund & The Student PIRGs, “Fixing the Broken Textbook Market: How Students Respond to High Textbook Costs and Demand Alternatives,” January 27, 2014, https://uspirg.org/sites/pirg/files/reports/NATIONAL%20Fixing%20Broken%20Textbooks%20Report1.pdf.

[3] “What is TAP,” University of South Florida Libraries, Accessed January 3, 2022, https://tap.usf.edu/what-is-tap/.

[4] “Evaluating Sources,” Australian National University Library, 2020, https://libguides.anu.edu.au/c.php?g=906019&p=6594267.

[5] Ryan McCready, “5 Ways Writers Use Misleading Graphs to Manipulate You,” Venngage, April 17, 2020, https://venngage.com/blog/misleading-graphs/.