How to make an effective chart? If this list of best practices is not exhaustive, it helps you already aware of the improvements you can make to your future graphic constructions.
1 – The message you want to convey through your chart shall become title conclusion. Use a descriptive subtitle if necessary.
2 – « Apply a low amount of information/ink ratio to give efficacy to your graph. » That’s what Edward Tufte teaches, one of the fathers of data visualization.
So, no frills in your chart! Your message is the most important element.
3 – Obviously, the choice of the type of chart to image your data cannot be due to chance. The graphics fall generally into four categories depending on what we are trying to highlight. It could be a comparison, a relationship, a distribution or a decomposition.
A graphical representation appropriate, it’s a shortened decision making time.
4 – The number of data sets must not interfere with the readability of your chart.
For example, it is best to be limited to three or four data series in a line chart. Beyond, this is more of the «spaghetti chart» For what is a pie, otherwise called « camembert », it is best to be limited to a maximum of five slices properly positioned. It is also used in data visualization, because the man knows not fully appreciate both the angles, areas or arcs of a circle.
The work of synthesis of Dona Wong, The Wall Street Journal – Guide to information graphics, is an essential reference in the field.
5 – Apply good practices on the construction of a graph, it is to understand the rules of visual perception of the human eye..
Among these rules, there are including the choice of colors. A set of monochrome colors with color differentiating for the key element, is much more effective to use all the colors of the Rainbow.
Similarly, a good appreciation of the contrast colors is paramount to allow your graph to be read without difficulty during a presentation.
6 – The size of the font, positioning and orientation of the text are also important options that promote a quick reading of the graph.
For example, it is best to write the title of the axis horizontally to avoid having to tilt the head during playback.
7 – It is important to not to distort information by misrepresented proportions, an inappropriate scale or a stretched graphic.
Edward Tufte measure this distortion through a coefficient of lie which is the relationship between the importance of the Visual effect and the importance of the qualitative gap. A close ratio of 1 is a sign of a chart that reflects the data.
8 – The use of a 3-d chart is to be avoided. The fact of introducing this 3rd dimension is the depth to give the volume does not allow the human eye to properly evaluate the information.
Indeed, even if this gives you a sense of aestheticism, this option ultimately harming the interpretation of your message and just degrade the lie factor.
9 – A chart reading course will naturally from top to bottom. In this case the legend for the chart should always be above it to limit the Visual scanning of the area displayed by the reader.
10 – Finally, to support the set of best practices, we can conclude with the fact that, over the time of assimilation is short more graph is effective..
During a presentation, the average time of attention given to the understanding of a slide is about five seconds… Thus, we understand the importance of the construction of a graph, and how nothing must be left to chance.
We integrated the maximum of these good practices into the Data Story Web application design, facilitate the grip and allow you to concentrate primarily on the effectiveness of your messages.
We have translated some parts or all of this page through a machine translation software. Some translations may therefore prove to be erroneous. Nevertheless, we hope that you find this information useful. Your comments are welcome. You can submit them using the Live Chat at the bottom of this page.