Creating Attractive and Effective Figures for Your Academic Paper

“A picture is worth a thousand words”, as the old saying goes. Your next academic research article will stand or fall with readers based on your images in your manuscript. Your figures can determine how well you sell your results to other academic researchers, peer reviewers, or a grant funding agency.

Start with your figures

One of the most common queries we field from researchers is how to start writing a paper. 

The best way to do this is by making figures. 

You’ve got your results: You collected data, analyzed it, and performed statistical analyses. Now you can draw graphs, make tables, or plot out some of your data. These form the basis of your results section which is almost always the first part of a paper to be written.

Each figure you create can then be written about and expanded. You can talk about the main outcomes, the key results, and then give context in a series of sentences grouped together into a single paragraph per figure.

Do not interpret the figures just yet. That comes in the discussion part of your manuscript. Stick to just the results and ‘nothing but the results’. But figures are key to building out your paper from this section. 

Don’t throw away unused figures

You might not end up using all the figures you make in your paper, but you will in other pieces of work, like posters and conference presentations, for example. 

One key rule of thumb in writing and article creation is never throw anything away. Ever. AJE has a fantastic figure and illustration creation service which you can learn more about here.

Before a journal reviewer or colleague even begins reading your paper, they have formed an opinion about the quality of your work. Your figures reflect your overall effort in experimental design, technical execution, and attention to detail.

Writing captions for your figures

Before starting to make your figures, consider your captions. Keep in mind that researchers will not read the whole of your study; rather, they’ll download a PDF or view your paper on their phone and click into the figures one-by-one. This means that you need effective, standalone figure captions that explain exactly that is being talked about in each image. Your figure captions need to be clear and self-explanatory outside of the context of the paper itself.

Example

Fig. 1: Image of liver during normothermic machine perfusion.

Caption: The hepatic artery (HA), portal vein (PV), inferior vena cava (IVC) and common bile duct (CBD) are all cannulated. The gallbladder (GB) is also present although this was often removed during the retrieval process before NMP. This image has been used with consent from the family of the donor.

Note that all necessary information has been included in this figure caption: 

• A short title that summarizes the image
• Clear labels
• Associated acronyms

Include enough information for a reader to just look at this and nothing else.

High quality figures begin in the lab

You need a plan before starting to conduct a piece of research. An analysis. A question you plan to address. 

Well-planned research means figures just ‘pop out’: It’s always going to be clear what kind of illustrations you’ll need to create: Histograms, Box Plots, Line Graphs, Scatter Plots. It doesn’t matter: You need to understand the difference between these and use them effectively in your research. Our handy video webinar explains more.

Use your muse. Think in terms of your controls and the kind of plots you’ll create. Ensure your colors are effective and that your key ideas come across clearly when others view your work.

Types of figures

Basic data presentation types

Bar graphs

One of the most common kinds of figures produced in academic research articles is called a ‘bar graph’. As shown above, these illustrate normally distributed data with error bars and a range of statistical significance. 

Box plots

If you don’t have normally distributed data (click here to learn the difference), then you need to create what’s called a ‘box plot’ where you present a median and interquartile range (IQR) instead of error bars.

Scatter plots

A final kind of basic data presentation is a scatter plot: In these, data are plotted across a baseline to clearly and transparently reveal distribution. 

Figures and colors

Color palette choice is very important in such presentationsThings to consider when choosing color for your graphs and tables:

• Who will be reading your article? 
• What kind of conclusions do you want others to draw? 

Avoid desk rejection

Indeed, when presenting any kind of data: Consider your color palette, as well as your scales, trend lines, legends, and labeled axes with units. 

Sounds basic? Well, these are reasons for article rejection. You’d be amazed how many papers get submitted that do not conform to these basic requirements. Check, check, and then check again. Or talk to one of our AJE experts by clicking here.

It’s key to choose a plot type that conveys your message in the simplest and most accurate manner. This could be a bar chart, a scatter or line plot, as we’ve discussed. Show your figures to colleagues and check whether their interpretation of your data matches your intended message. And check our Rules for figure creation, below.

Rules for figure creation

Rule 1: Know your audience.

A figure speaks to your readers. Think about your message and what you want your figure to report before putting it together.

Rule 2: Identify your message.

A clear message in a figure is tantamount. Message over beauty as the man said. When you decide which and how many figures you want to show in your paper, decide on one message that you want to communicate with each. This message should ideally represent one of your conclusions or parts of it.

Rule 3: Work at actual publication size when putting together your figures.

Journals have different guidelines, page sizes and formatting requirements. Have a journal in mind before starting to create a figure. Think about your journal guidelines and print and measure to ensure a good fit when a reader sees the final work. Readers are the most important people in the whole publication process after all.

Rule 4: Use the correct size and type of font and keep a sense of scale.

Adapt to your medium. Is your figure a 3D PDF or will it be printed in black and white in the journal.

Rule 5: Don’t clutter.

It’s confusing for readers. The simpler your plot, the easier and quicker your reader will grasp it. In some data visualization programs, the default options are full of clutter: remove everything you can from your plot while still telling the same message.

That’s it! Remember the five key rules for figure creation.

Final thoughts

Whatever your figure shows, there are three key points: 

• Keep the message clear
• Keep it simple
• Following best practices in terms of data presentation

Don’t hesitate to try something new.