Digestive Disease Week® (DDW) 2020 abstract submission is now open. Our next post in the series How to Write a Good Abstract focuses on creating high-quality figures. Guest author Kira L. Newman, MD, PhD, walks you through the steps.
As a reminder, authors can submit up to two images per abstract, which may need to be scaled for publication. To see examples of DDW journal publication layout, view the DDW 2019 abstract supplements published in Gastroenterology and GIE.
Have questions on figures that aren’t answered here? Write them in the comments below!
High-quality figures are part of making a successful impression with your abstract. Quality is comprised of three major categories:
- Technical adequacy
- Conceptual clarity
- Appealing design
The figure should meet size, file format and resolution requirements. When in doubt, go for a higher resolution (usually measured in dpi). Tiffs are the best file format for maintaining resolution. If you’re including text in your figure, stick with 600 dpi resolution. For line-drawings, this goes up to 1200 dpi. If you have to resize a figure or image, save the original and the adjusted images as two separate files. That way you always have a back-up.
Think carefully about what you want to communicate and choose a figure type that helps illustrate this using your data. Are you presenting comparisons? Distributions? Temporal trends? Once you know what you want to show, pick a software that can create high-quality figures. This includes R, Excel, GraphPad and others. R is one of the most versatile options but has a steep learning curve. Fortunately, there is lots of sample code available online.
Your figure should also fit into the larger message of your abstract. Use figures to illustrate concepts and findings. In the word count-constrained word of abstracts, a figure is guilt-free pleasure; figures (and tables) don’t count towards your word count.
Looking at publications with similar types of figures to those you want can help you refine the design. Many journals provide style guides for figures that can help, too. Design issues to attend to include color palate, font type and size, line weight and label locations.
General best practices include using colors that are contrasting in both color and grayscale, using a sans-serif font like Arial or Helvetica, avoiding bold and italic type, and eliminating unnecessary visual clutter, like background grids.
“How to Create Publication-Quality Figures: A step-by-step guide (using free software!)” by Dr. Benjamin Nanes has detailed instructions and includes code.
“Data to Viz” is an interactive site that can help users select the best type of visualization based on their data. It also includes links to code for creating different types of figures in a variety of free softwares.
“Data Visualization” from Duke University Libraries has a detailed list of visualization types, including some more creative options, and a list of the tools needed to create each type.
“Data Visualization” from the University of California Berkeley Library has helpful information on design considerations and tools.
“R” and R studio are statistical computing software that is extremely versatile for making reproducible, publication-quality figures. The learning curve is steep, but if you had one software for making figures, this might just be it. There are many tutorials and examples of code online. Free and available for Mac, Windows, and Linux.
“+Datavisualization.ch” has a curated list of links to data visualization tools online.