Sport Informatics and Analytics/Audiences and Messages/Visualising Data

Introduction
At some point in the sport analytics process we share our findings with an audience. In doing so, we make decisions about what we share and how we share it.

In 1919, Willard Brinton published his guidelines for graphic methods to share facts. The first chapter of the book includes these observations: After a person has collected data and studied a proposition with great care so that his own mind is made up ... The larger and more difficult part of the work is to convince the minds of others that the proposed solution is the best one - that all the recommendations are really necessary.

Wolfgang Iser (1976) suggested that when we produce a story to share we should think carefully about how we construct the story and imagine the recipients of the story. He notes that any story has "a network of response-inviting structures" (our emphasis) that enable the reader, listener or viewer "to grasp the text".

The availability of digital platforms to share audio-visual resources has extended the reach of such stories.

Maria Popova (2009) has looked at the impact digital platforms have had on the way data are shared. She notes that at "the intersection of art and algorithm": Ultimately, data visualization is more than complex software or the prettying up of spreadsheets. It's not innovation for the sake of innovation. It's about the most ancient of social rituals: storytelling. It's about telling the story locked in the data differently, more engagingly, in a way that draws us in, makes our eyes open a little wider and our jaw drop ever so slightly. And as we process it, it can sometimes change our perspective altogether. Michael Friendly provides an insight into the social ritual of storytelling with his account of the emergence of a 'golden age' of statistical graphics in the second half of the nineteenth century.

Robert Kosara (2007) noted that two cultures exist in visualisation: technical, analysis-oriented (pragmatic visualisation); and artistic pieces. He suggested that a third culture could be created by the interchanges between the existing two cultures. Robert proposed that the third culture: not only takes ideas from both artistic and pragmatic visualization, but unifies them through the common concepts of critical thinking and criticism. Visualization criticism can be applied to both artistic and pragmatic visualization, and will help to develop the tools to build a bridge between them. Robert identified three characteristics of visualisation that would inform discussions about any visualisation:
 * It is based on (non-visual) data.
 * It produces an image.
 * The results are readable and recognisable.

Visualisation
The mind map for this topic shares a variety of links to data visualisation and storytelling.

The tools we have to share our data with audiences are available in open formats and as commercial products. (An example of the use of an open format is shared in the use of R in this course.) You will be sharing your data stories in what Elijah Meeks (2018) suggests is a third wave of data analysis and data exploration in which there is an emphasis on design and impact that have a pedagogical dimension (Michael Correll, 2018).

Trina Chiasson, Dyanna Gregory and their colleagues (2014) have provided a comprehensive guide to preparing and visualising information. They have shared the source code for their work on Github. They note: Data come in all different shapes, sizes, and flavors. There’s no one-size-fits-all solution to collecting, understanding, and visualizing information. Some people spend years studying the topic through statistics, mathematics, design, and computer science. And many people want a bit of extra help getting started. The awareness that "no-one-size-fits-all" has led to some fascinating discussions about the aesthetics of visualisation. Ben Schneiderman (1996), Leland Wilkinson (2005) , Howard Wainer (2005) , Stephen Few (2013) , David McCandless (2016) , Alberto Cairo (2012, 2016) , Gregor Aisch (2016) , Giorgia Lupi (2016) , Kieran Healy (2018) and Winston Chang (2019) , among others, have explored the ways in which we visualise data stories.

There is more information about visualisation discussions at this location. Some of the literature on visualisation can be found in this resource.

If you would like to consider how a minimalist approach to data might enhance the impact of your visualisations, you might like to have a look at Joey Cherdarchuk's discussions of: bar charts, data tables , pie charts and plotting distributions.

Payman Tei (2017) and (2018) offers some insights into the characteristics of effective and engaging data visualisations that convey information and share stories. Lisa Rost (2017) discusses three characteristics of effective charts that can be generalised to other visualisations: what is your point?; how can you emphasisze your point?; what does your visualisation show exactly? Giorgia Lupi (2017) invited us to consider these kinds of questions in the stories we share through visualisation of data as a placeholder in our conversations.

Enrico Bertini (2017) discussed the use of visualisation in the process of iterative data analysis.

Elijah Meeks (2017) discussed gestalt principles for data visualisation. He noted "it is critical to understand gestalt when you are creating more complex data visualization products like network visualization or hierarchical diagrams". These principles include: similarity, proximity and enclosure; common fate, parallelism and connectedness; network visualisation; and figure/ground and metastability. Subsequently (Meeks, 2018), he wrote about a third wave of data visualisation that acknowledged "convergence isn’t just happening in the capabilities of tools but also in the expectation of users".

Richard Brath and Ebad Banissi (2016) have discussed in detail the use of typography to expand the design space of data visualisation.

Yan Holtz and Conor Healy (2018) shared a decision tree to identify the graphic possibilities for different kinds of data: numeric; categoric; numeric and categoric; maps; network; and time series. They provide a range of caveats about data presentation.

Andy Kirk (2019) discussed opportunities to learn about visualisation in a range of sports. He shared his observations about "techniques for visually portraying trends, changes and activities over time".

Senthil Natarajan (2019) discussed opportunities to learn about visualisation in a range of sports. He suggested: Cross-pollination isn’t just about sharing your own work, though. It’s also about an entire community coming together to share resources.

Graphical Perception
William Cleveland and Robert McGill (1984) sought to provide a scientific foundation for the use of graphical methods for data analysis and data presentation. Their approach was based on graphical perception, "the visual decoding of information encoded on graphs".

William and Robert identify ten elementary perceptual tasks that people use to extract quantitative information from graphs:
 * Position (Common Scale)
 * Position (Non-Aligned Scales)
 * Length
 * Direction
 * Angle
 * Area
 * Volume
 * Curvature
 * Shading
 * Color saturation

They conclude: The ordering of perceptual tasks does not provide a complete description for how to make a graph. Rather, it provides a set of guidelines that must be used with judgement in designing a graph. Many other factors, such as what functions of the data to plot, must be taken into account in the design of a graph.

Wolfgang Aigner (2010) and Colin Ware (2012), among others, has extended this work to explore in detail the science of perception in the context of visualisation. See also, the Massachusetts (Massive) Visualization (MASSVIS) Dataset that seeks to understand the cognitive and perceptual processing of a visualization. is essential for effective data presentation and communication to the viewer. The MASSVIS Database was constructed "to gain deeper insight into the elements of a visualization that affect its memorability, recognition, recall, and comprehension". See Lee Sukwon, Kim Sung-Hee and Hung Ya-Hsin's (2016) discussion of making sense of unfamiliar visualisations as a contribution to the discussion of perception and sense making. Zdenek Hynek (2016) discussed the bandwidth of perception.

Examples
You will have many choices available to you as you explore your visualisation options. We share three examples here.

Voronoi diagrams
You might like to have a look at this discussion of one visualisation method, a Voronoi diagram, to discover how it has been used in a variety of sport contexts.

ggplot2
One of the topics in this course is Using R. Within the R platform, you have a variety of options to visualise data. One of these options is the R package ggplot2. Kieran Healey (2017) shared a practical introduction to R and ggplot2. Alboukadel Kassambara provides a detailed introduction to the use of the ggplot2 package. We suggest that if you are thinking of using ggplot2 you look at Alboukadel's ggplot2 Essentials to inform your visualisation journey and the STHDA guide. See also, Baptiste Auguié's (2018) guide to sharing multiple plots. For an example of an author working through the use of ggplot2 with sport data, see Alice Sweeting (2019).

Hasse Walum and Desiree de Leon (2019) have provided a novel approach to learning ggplot2. Their guide includes ways to customise plots.

Histograms
Aran Lunzer and Amelia McNamara (2017) provided an insight into the construction of interactive histograms.

We recommend that you read their essay. It involves some interactive features that encourage reflection on our audiences and messages theme. The interactive visualisations in the essay include some NBA and New York Marathon data.

Their conclusion offers some background information about how they constructed their essay: This essay is built using styling and tooltip behavior from Bootstrap, with scrolly responsiveness derived from Jim Vallandingham's (2015) So You Want to Build A Scroller. The interactive visualizations make extensive use of D3.js, and utility functions from lively.lang. While the code was primarily generated to power this essay, it is available on GitHub.

Circular plots
Andrew Howe (2018) used a circular plot to create an interactive visualisation for goal scoring in the 2017-2018 WLeague association football competition in Australia. The visualisation has data on goal scoring that goes back to the origin of the WLeague competition in 2008-2009. Andrew's use of circular plots was informed by Nikola Sander and her colleagues' (2014) visualisation of migration flow data. There is an R package, circlize, that provides an example of how to develop a circular layout. Zuguang Gu (2017) provided a tutorial for circular visualisation in R.