Instant Ideas and Collaboration
#uklibchat will be on Tuesday 1 December from 6:30-8:30pm GMT time. The topic will focus primarily on how libraries can effectively communicate different types of data, whether through visual means such as infographics, charts or images.
We have a Data: The Big, The Small, The Beautiful chat agenda where you can add your questions for discussion on the night. Join us to share and exchange ideas and discuss data visualisation and data communication. Use the hashtag #uklibchat to contribute to the discussion.
Our featured post is written by Sarah Stewart @biostew, a recent member of #uklibchat team.
Data are vital, ubiquitous parts of the information landscape – not only do information professionals in all sectors collect, store and manage data in some form, but they may also need to communicate it to others, whether stakeholders such as funding bodies or managers, or users. Data can be harnessed to drive decision-making and development in libraries, whether developing collections, services or investigating user experience, and can also be used to demonstrate impact and value. While these data analytics or measures are important, communicating them effectively is an often overlooked part of the process.
As a former researcher in the biosciences, with a strong interest in the history of science, I was fascinated by some of the famous data visualisations, such as Florence Nightingale’s 1858 ‘coxcomb’ diagrams showing causes of mortality in the Crimean War, Darwin’s 1859 diagrams of the Tree of Life from his work, On Origin of Species, and Mendeleev’s Periodic Table of the Elements. These charts and diagrams communicate data in a way that is universally understandable, and their visual impact has stood the test of time.
Data visualisation provides a graphical means of communicating data clearly and efficiently to users, whether these graphical images are statistical graphs or plots, charts or tables, infographics or even maps. Data visualisation ideally should make the data ‘speak for itself’ and tell a story, whether this is a comparison or an attempt to understand the underlying causality of a phenomenon. The design of the data visualisation should ideally follow from the question being asked.
‘Big Data’ has become a ubiquitous term to describe the vast and rapidly-changing volumes of data generated from a myriad of activities. Even applications such as Twitter may produce huge amounts of data to be stored, used and re-used through search and discovery. While the volume of data may seem overwhelming, information professionals are well-placed to ‘master the art’ of data communication, which can be accomplished effectively through data visualisation. What types of Big Data, if any, are being used in libraries or archives and other memory institutions (such as museums or art galleries)? ‘Small Data’ – the individual datasets and spreadsheets held on one computer terminal can also be important and can strongly influence and advocate for development if employed correctly.
Data visualisation provides a great opportunity for librarians and information professionals to dig deeper into their data and use it to tell stories, as data such as that held in a spreadsheet becomes quickly understandable and patterns and inferences can be teased out. This type of communication can provide multiple opportunities to showcase and promote new services, develop collections through alerting users to new possibilities such as new electronic resources or gain valuable feedback which can be used to further various developments across the institution. In July, 2014, the Natural History Museum launched its Data Portal which provides open access to specimen data, including some published datasets authored by Museum Scientists. The Data Portal includes some spectacular visualisations. Specimens are mapped out and are discoverable by geographic region, and timelines of historic collections which look at species distribution can also be found and mapped. The Data Portal provides a spectacular reference for anyone researching Natural History collections, or even a single specimen held at the Natural History Museum. Similar work could be done in other memory institutions such as libraries.
While data visualisation can provide a lot of opportunity, it can also be challenging. In order to effectively visualise, or communicate data, it must first be understood, which might be difficult with complex and dynamic datasets. Some knowledge of statistics, graphic design and computer programming may be required in order to produce professional-looking infographics, charts and maps. There may be some ethical challenges in addition to analytical challenges, particularly for sensitive or personal data. Finally, the technical tools such as software required to produce data visualisations may not be readily available, though Open Source software such as Tableau and Processing, a Java-based computer programming language for producing visualisations are freely available and free to download on-line. Finally, making the data ‘speak for itself’ may in itself be a challenge. While data visualisation can bring forth patterns not immediately obvious from a raw dataset, it can also obscure or confuse. Best practice is therefore very important when visualising data.
1.) Let the data dictate the type of visualisation – Communicate key aspects of the data in an intuitive way. Comparisons can be shown quickly and effectively with bar charts, geographical plots can be visualised on maps.
2.) Know your audience – Who is the data visualisation being prepared for? Managers? Library Users? Try to put the data into context.
3.) Design graphics that can stand alone outside of the context of any report. The data visualisation should be a ‘stand-alone’ summary of the key points drawn from the data.
4.) Keep it simple – Simple, minimal graphical visualisations are often the most effective. Do not overload with superfluous text and labels, and keep the format minimal. Do not let ‘showy’ graphical design techniques dominate and distract from the data itself!
#uklibchat Collaborative Data Visualisation Experiment – Hand’s On!
The best way to learn is to practice, so following the chat on Tuesday; I will make the raw data files from the Twitter analytics available. We would be really interested in seeing what you are able to create from the data, and we will display your work in a second feature blog post ‘gallery’. Please send any completed data visualisations to firstname.lastname@example.org by December 31st. We look forward to seeing your own data creations!
Sarah Stewart (@Biostew), a recent member of the #uklibchat team and Information Assistant at the Natural History Museum Library & Archives in London (@NHM_Library). Sarah has previous experience in the biological sciences and a strong professional interest in all aspects of data.