Visual displays should be used to directly support the text and provide a visual representation of what is being discussed. They can break up long slabs of text and create visual appeal for a publication or website, providing an easy entry point for users. They should be relevant to, and referenced in, the text.
The 2 common forms of visual display are tables and figures. The term ‘figure’ is simply defined as a visual display that is not a table. Some visual displays – such as maps, photographs, infographics or videos – may be produced as products in their own right, but most are incorporated into a document, publication, website or presentation.
Reduce clutter
It is important that the message or data shines through your visual display. It should therefore be the most prominent and legible item in your display (Kirk 2012). Reducing clutter can help to achieve this.
Remove citations, caveats, logos, background shading, borders, and other nonrelevant data or graphic elements within the table or figure, where possible.
Reduce ‘ink on the page’ and pay attention to the data:ink ratio (Tufte 2001). This means:
- reducing the nondata ink by using paler tints or finer lines for nondata elements, such as grid lines, arrows, borders and shading
- increasing the data ink by making key points bold, highlighted, coloured, underlined and so on (Few 2012).
Use colour wisely
Be restrained with the use of colour. It can be more powerful and effective to use a single splash of a strong colour to highlight a key point than to colour the entire table or figure.
When applying colour, be aware that some colours can carry cultural meaning, such as green for good, go or environment; and red for bad, stop, warning or heat. Colours also create a psychological response – bright colours are exciting, energetic and attract attention, whereas pale colours are soothing and receding.
Look for opportunities to use colour meaningfully, and be careful not to imply meaning where there is none. For example, you could use 2 shades of a single colour to indicate different years of data, rather than red and green, unless you want to imply a judgement about the data series.
Ensure appropriate sizing
Resize or redraw the table or figure to fit the layout of the publication, ensuring that critical parts of the figure are presented at an appropriate scale for legibility.
Ensure that any text is legible. If possible, use Arial or Helvetica at 9 pt (minimum 7 pt) at the size the figure will be used in the layout (note that, if you resize the figure in the layout, the text will also scale). Text can be rotated, preferably to 30°, 45° or 90°, but never set vertically (i.e. with letters the right way up but set underneath each other):
Graph showing that text can be rotated but never set vertically
Ensure consistency
Be consistent with legend use and styling for all tables, graphs and other visual displays throughout your publication.
Make sure that comparable information is displayed in equivalent formats (e.g. do not show one year’s dataset in a line graph and the same dataset for a different year in a bar graph). Graphs that are to be compared with each other should also have comparable scales and categories, and consistent styling.
Ensure understanding and comparability
Standardise common elements across similar tables and figures (e.g. appearance of legends, data elements, axes, typography, borders, colours, line styles).
Consider adding direct labels or a contrasting colour to highlight key points in figures if this might assist user comprehension of the message:
Graph showing use of colour to highlight a key data point
Use the text to point to the key messages
It is good practice to discuss the key points of a table or graph and then direct the user with an in-text reference, rather than just announcing the inclusion of the table or graph. This highlights the message directly, and makes the text more active and interesting:
Australian house sizes have increased rapidly over the past 50 years, with Sydney showing the greatest increase in average house size (Table 1.1).
not
Table 1.1 shows the average house sizes for Australia’s capital cities from 1960 to 2010.
In some cases, it can be appropriate to use the name of the table or graph itself to highlight the key message. This is not generally done in academic journals or more formal publications, but can be a valuable technique in many publications, particularly for general audiences.
Prepare tables and figures for publication
If you are preparing visual material yourself, ensure that graphs and other figures are of high resolution. Where possible, produce images in a format that is scalable for different devices, such as smartphones, desktop and print (e.g. scalable vector graphics: .svg).
If a graphic or digital designer is producing your visual material, give them all your content in editable formats, such as scalable vector graphics (SVG), encapsulated postscript (EPS) or Adobe Illustrator (AI) files.
If the display is a data visualisation, provide your designer with the data that will underlie the graphic. Data should be provided in an editable and ‘open’ format, such as comma separated value (CSV) or Excel (XLS) spreadsheet. This will allow the designer to manipulate the data into a format that can be easily used by the graphing or design software that will produce your display.