Dr. Avriel Licciardi, Research Communications Strategist
July 2022Figures are an extremely important aspect of any scientific publication. In fact, many readers will only view the figures and abstract before deciding to (or not to) read your paper. Thus, it is essential to have well-constructed figures that are clear, accurately display your data, and have excellent visual appeal.
All figures should convey information from data sets both quickly and effectively. However, it is common that these graphics, which aim to explain and highlight scientific research, confuse the reader. A confusing or inadequate figure may be due to multiple reasons, including:
• data errors or misleading representation
• missing or incorrect information (e.g., no axes titles, units, scales)
• poor color choices and/or unnecessary visual elements
• overlapping text, lines and/or images
• small and/or inconsistent font size
• inadequate formatting (e.g., blurry or low-resolution images)
• the absence of an appropriate and helpful figure legend, caption or explanation
• missing or incorrect information (e.g., no axes titles, units, scales)
• poor color choices and/or unnecessary visual elements
• overlapping text, lines and/or images
• small and/or inconsistent font size
• inadequate formatting (e.g., blurry or low-resolution images)
• the absence of an appropriate and helpful figure legend, caption or explanation
Because figures come in an extremely wide variety of types and formats, we provide general figure-making guidelines that will transform any figure into a scholarly graphic worthy of publication.
Figure Preparation Guidelines:
1. Read your target journal’s instructions for figures closely. Most journals provide information on the number of tables and figures you can include, numbering style, titles, image resolution, and file formats, among other specifications.
2. Know your audience. Make sure that your figure correctly displays the data and conveys all the relevant information to the appropriate audience. For example, student audiences may require extra information to fully explain a concept compared with established scientists.
3. Identify the role of each figure in your manuscript. Specifically, what is the underlying message of the figure and what type of plot/graph can best express your data? Once you identify the reasoning for a figure, it serves as a guide for the figure’s design.
4. Use the right figure-making tool. Depending on the type of visual you are trying to create, there is often a dedicated or commonly used tool that will do exactly what you want to achieve. If you are unsure of the best tool to use, ask your colleagues, advisors or mentors, or refer to the methods sections and figures in papers within your research field.
5. Present your data comprehensively. Each piece of data has a description, a number, a unit, and an uncertainty. When plotting data, it is best to place all four components of the data in the figure. For example, if any data points are removed, you must explain why. If error bars are present, explain what they represent.
6. Present your data as objectively as possible. A scientific figure should illustrate your data as objectively as possible to not mislead the reader. Use the simplest type of plots that convey your data accurately, and only include relevant labels and values.
7. Use color when it enhances your graphic (especially since most articles are now read online), but make sure that no information is lost when the figure is printed in grayscale.
8. Make your data stand out. Avoid all unnecessary or confusing visual elements such as too many colors and labels, useless grid lines, and gratuitously colored backgrounds.
9. Be consistent and clear. Present values and details consistently in your figures, as well as in your tables and text (e.g., abbreviations, acronyms, treatment names). Write clear, informative titles for your figures, and label any axes, panels, and legend components.
10. Make sure that every element of the figure is fully explained. If the explanation is not on the figure or in its caption, then the explanation must be in the main text. Each figure in your manuscript should be self-explanatory and understood independent of the text.
11. Write effective figure captions. The figure caption should make it possible for your reader to interpret and understand the significance of the figure without referring to the main text.
2. Know your audience. Make sure that your figure correctly displays the data and conveys all the relevant information to the appropriate audience. For example, student audiences may require extra information to fully explain a concept compared with established scientists.
3. Identify the role of each figure in your manuscript. Specifically, what is the underlying message of the figure and what type of plot/graph can best express your data? Once you identify the reasoning for a figure, it serves as a guide for the figure’s design.
4. Use the right figure-making tool. Depending on the type of visual you are trying to create, there is often a dedicated or commonly used tool that will do exactly what you want to achieve. If you are unsure of the best tool to use, ask your colleagues, advisors or mentors, or refer to the methods sections and figures in papers within your research field.
5. Present your data comprehensively. Each piece of data has a description, a number, a unit, and an uncertainty. When plotting data, it is best to place all four components of the data in the figure. For example, if any data points are removed, you must explain why. If error bars are present, explain what they represent.
6. Present your data as objectively as possible. A scientific figure should illustrate your data as objectively as possible to not mislead the reader. Use the simplest type of plots that convey your data accurately, and only include relevant labels and values.
7. Use color when it enhances your graphic (especially since most articles are now read online), but make sure that no information is lost when the figure is printed in grayscale.
8. Make your data stand out. Avoid all unnecessary or confusing visual elements such as too many colors and labels, useless grid lines, and gratuitously colored backgrounds.
9. Be consistent and clear. Present values and details consistently in your figures, as well as in your tables and text (e.g., abbreviations, acronyms, treatment names). Write clear, informative titles for your figures, and label any axes, panels, and legend components.
10. Make sure that every element of the figure is fully explained. If the explanation is not on the figure or in its caption, then the explanation must be in the main text. Each figure in your manuscript should be self-explanatory and understood independent of the text.
11. Write effective figure captions. The figure caption should make it possible for your reader to interpret and understand the significance of the figure without referring to the main text.
Additional Tips:
• It is a good idea to make your figures before writing your manuscript. This way you can save text (the message is already in your figure!), have a graphic to reference while writing, and organize your thoughts ahead of the writing process.
• In many fields it is customary to end with a “summary figure” of your results and how your data compares to other related data/publications.
• Make sure to reference all your figures in the main text.
• It is best to develop figures in a vector format so that they can be resized by your target journal. Invest in learning software that uses vector graphics, such as (but not limited to) Adobe Illustrator.
• Do not repeat the contents of your figures within the main text. Instead, use the text to focus on the significance or key points of the data displayed in your figures.
• If you are including a figure that has already been published, acknowledge the source.
• Before submission, print out your paper to confirm that none of your figure labels are too small. This quick check can avoid your manuscript being immediately returned to you due to a “technical failure.” Many journals have specific limits on figure text size.
• In many fields it is customary to end with a “summary figure” of your results and how your data compares to other related data/publications.
• Make sure to reference all your figures in the main text.
• It is best to develop figures in a vector format so that they can be resized by your target journal. Invest in learning software that uses vector graphics, such as (but not limited to) Adobe Illustrator.
• Do not repeat the contents of your figures within the main text. Instead, use the text to focus on the significance or key points of the data displayed in your figures.
• If you are including a figure that has already been published, acknowledge the source.
• Before submission, print out your paper to confirm that none of your figure labels are too small. This quick check can avoid your manuscript being immediately returned to you due to a “technical failure.” Many journals have specific limits on figure text size.
Previous Article