Story Review Formatreport review format
Telling a story with dates
How does a graphic artist tell a story with a visualisation? Analyze must find the history that will support the dates. This is something traditionally done in the field of journals, and it has become very good in the field of history with visualisation through information graphics. With this in mind, here are some editorial strategy for narrating a good story that also applies to visualizing it.
Identifying the story line helps you determine if you actually have a story to tell. But if not, this visualisation should perhaps rather be used to aid explorative EDA than to provide information. For the explorative visualisation creator, however, it is still important to stimulate the viewer's fantasy, examine the relationship between the information and make it easier to interact with it - think gameeification.
Visualisation must be surrounded by the information layer of the public, right and wrong: be impartial and provide well-balanced. Visualisation should be free of distortions. And even if it argues to exert pressure, it should be founded on what the information says, not what it should say.
Bufte found many diagrams that mislead the viewer about the basic information and developed a calculation to quantitatively represent such a deceptive graph, the "lie-factor". "The lie ratio corresponds to the magnitude of the effect shown in the graph by the magnitude of the effect in the graph.
You can easily promote objectivity: label to prevent ambiguities, adjust graphical sizes to meet your measurements, use standard entities, and prevent drift. Equilibrium can result from alternate visualizations (multiple clusterings; trust interval instead of line; alternating time bars; alternate colour ranges and allocations; adjustable scaling) of the results in the same render.
Spectators and decision-makers will ultimately find discrepancies, which in turn will lead to the loss of confidence and reliability of the designers, no matter how good the story is. Do not be discriminative about the information you are including or excluding unless you are sure that you are giving your audiences the best presentation of what the information "says".
These selectivities include the use of discreet scores when the dates are continual, handling absent, outliers, and out-of-range scores, any time bands, caps, volumes, scopes, and interval. The viewer will finally find out and loose confidence in the visualisation (and everyone else you produce). Also make sure to really declare the dates, not just decorating them.
Don't drop into a "it looks cool" setup if it may not be the best way to clarify the datas. If you spend more of your journalist and writer work on and improve your visualizations than you do on them, you're probably doing the right thing.