[MUSIC] Journalists are increasingly integrating visualization into their narratives. In some cases, visualization is functioning like a written story. To captivate readers by drawing them into the story with greater detail. The Oxford English Dictionary defines narrative as "an account of a series of events, facts, etc. given in order and with the establishing of connections between them." Narrative visualization is often used in journalism contexts, Which tends to include aspects of both exploratory and communicative visualization. Today we are specifically looking at how to do narrative storytelling with data. A general story model is an ordered sequence of steps with a clearly defined path through it. Each step can contain text, images, visualizations, or videos. Data journalism is different from traditional storytelling. For traditional journalism stories, journalists collect data through research, interviews, etc. to assemble key facts. Journalists then tie the data together to produce a narrative. But not all of the data were available for the readers. Data is secondary when it compares to the narrative. However, journalists who work on data- driven stories use visualizations for both the exploration/ analysis and the presentation. So the narrative is built around data. In traditional stories, order roughly corresponds with time, which is crucial to understand causality-- events that happened earlier can influence later events, but not the other way around. While data-driven visualizations can also be organized in a linear sequence, they can also be interactive. In response to the growing number of online visualizations designed to convey a story, Edward Segel and Jeffrey Heer published an article with the title "Narrative Visualization: Telling Stories with Data." They reviewed the background of storytelling through data and then in comparison with traditional storytelling. At the end they presented three ways to distinguish categories of narrative visualization. The first one is genres. The second one is the visual narrative tactics that direct attention, guide view transitions, and orient users. The third is the narrative structure tactic, such as ordering, interactivity, and messaging. All these provide a general framework that opens the discussion of narrative visualization to a wider range of examples. Segal and Heer analyzed visualizations from online journalism, blogs, instructional videos, and visualization research. They identified seven genres of narrative visualization: Magazine style, annotated chart, partitioned poster, flow chart, comic strip, slide show, and video. Genres are different based on the number of frames, which is the distinct visual scenes, and their ordering. The purpose of the genres is to decide which one is appropriate according to the situation. TV is more likely to use film, video, and animation, While during a business meeting, people were more likely to use annotated charts, flow charts, or slideshows. Visual narratives deals with the principles of how visual media can be organized to produce a narrative experience. For example, cultural factors affect how we read the text: from left to right or from right to left. There are also three ways affecting how we read a visualization. The first one is visual structuring. Visual structuring refers to mechanisms that communicate the overall structure of the narrative to the readers. These design strategies help orient the readers early on for establishing short, checklist, consistent visual platforms and allowing the readers to track their progress through a progress bar or timeline slider. Some visual techniques can further establish the order in which the eyes visit elements in the story. Preattentive visual features such as color, size, and orientation can do the highlighting. Gestalt grouping via features such as spatial proximity, containment, or connection can direct readers to first look at the grouped element. Vectorial reference, most commonly in the form of arrows, is also a powerful technique for directing attention. Transition guidance concerns techniques for moving within or between visual scenes without confusing the readers. A common technique from film is called continuity editing, though, there are other strategies such as animated transitions and camera motion. Sometimes data can be ambiguous and contextually deficient. Visualizations are subject to interpretation. Narrative helps reduce uncertainty, connect data with context, and describe a specific interpretation. Narrative is the key, and in the narrative framework of journalistic visualization there includes three components: ordering, interactivity, and messaging. Ordering refers to the ways of arranging the path readers take through the visualization. Sometimes this path is linear and prescribed by the author, and sometimes there's no path suggested at all, which opens for the readers for random access. And other times, the user must select a path among multiple alternatives, which we call user-directed. Interactivity refers to the different ways a user can manipulate a visualization-- such as filtering, selecting, searching, and navigating-- and also how the user learns the methods for explicit instructions and initial configurations. Messaging refers to the ways a visualization communicates observations and commentary to the readers. This might be achieved through short text fields such as using labels, captions, headlines, and annotations or more substantial descriptions such as an article or summary. Interaction is one of the most important parts of visualization. This allows the readers to not just see the data but quickly change the view or add different data. This makes the analysis much faster and more effective. Animation is one of the kinds of interactions. Animations can automatically direct the flow of the presentation. For example, the stock market has been fluctuating a lot in the past month. The stretching time axis as readers scroll down makes for an easy-to-follow visual cue. Usually, we would include animation to graphs that need to show trends and patterns. Among the attempts to add interaction to our story, games are perhaps the most interesting and certainly the most popular method to use. Games about creating goals for the user. Adding meaningful and ethical interactions as the format of the game can impact user behavior around the graphics. Features in an interactive visualization should have clear affordances to distinguish them from other states and the surrounding layout. For example, an enabled state communicates an interactive component or element. A disabled state communicates a non-interactive component. A hover state communicates when a user has placed a cursor above an interactive component. A selected state communicates a user choice. Interactions allows the change of parameters such as dragging a slider or allowing filtering or sorting. Sometimes it allows users to enter their own data to carry or easily undo a parameter change. A map makes a good example for adding navigation. It starts off with a broad overview, and then if there's visual information that the readers are curious about, such as a particular sub region, readers can zoom in and get additional information. We can also think of zooming as a form of filtering. In most journalistic visualizations, context is added through the use of labels, captions, or other annotations. Textual annotations integrated directly with a visualization can further guide a user's interactions-- emphasizing certain points, prioritizing particular interpretations, or preempting the user's curiosity on seeing a salient outlier or trend. There are two types of annotations: observational and additive. Observational annotations provide context by supporting refelction On a data value that is depicted in the visualization. This annotations facilitate comparisons and often highlights or emphasizes extreme values or outliers. For interactive graphics, observational annotations are sometimes revealed when hovering over visual elements. Observational annotation is the information that can be observed in the graph but is being reinforced through the use of text. On the other hand, additive annotations provide context that is external to the visual representation and not clearly depicted through the data. This is information that is relevant to the topic or the understanding of the data, such as providing background on the events or actions. So far I have just picked to talk about some forms of interactivities and messaging in detail, but there are actually a lot more that are used in visualization. Here are some more to show what other messaging and interactivity methods are usually used in narrative visualization. The yellow boxes refer to interactivity, and the green boxes refer to messaging. For example, this graphic allows you to slide the timeline so you can pick the year that you were born to see how the whites born in that same time have leaned politically over their lives. The slider also enables animated transitions so you can see the trends. Throughout the visualization, there are annotations and paragraphs to explain the story behind it. Similarly, this visualization also uses interactivity and messaging to show the number of jobs that have changed in a particular industry. When readers scroll down, more narrative will be provided. When the readers hover over a particular line, a smaller window will pop up and illustrate the details. There's also a progress bar on the right to show the status of the current slide in relation to the total number of slides in the entire new story. Captions, introductory text, and instructions are there to help readers understand the visualizations.
