It is common for authors and editors to spend a lot of time discussing and debating the graphical presentation of data, and that can be time well spent as it often yields a better final figure that meets the needs of the journal, the authors, and the readers. If a picture is worth a thousand words, then a well-designed figure can have two to three times that value. Over the past few months, the editorial office has been revising the guidance we provide to authors regarding figures. Hereinafter, you will find the section from the journal's Instructions for Authors relating to the new guidelines for figures. Although these graphical standards are not new, formalizing these guidelines as a document and including this information in the Instructions for Authors are new. We hope that making these standards available to authors as part of the Instructions for Authors will be helpful and can reduce the effort required to produce final publishable quality figures that conform to the journal's style guidelines.
Figures that do not conform to the journal style guidelines are not wrong, and deviations from these style guidelines in some cases may be advisable depending on an author's intentions and the story to be told by his/her data. Nevertheless, there are best practices that can help enhance the clarity of the author's message rather than detract from the story by creating unnecessary distractions for the reader.
In summary, there are a few key principles to bear in mind when developing figures for publication:
- Strive for clear, simple, and consistent style for all figure elements.
- Figures should stand alone, and figure legends should provide sufficient descriptions to allow readers to understand figures apart from the text.
As a way to make learning fun, I have prepared a few figures with multiple mistakes (Figs. 1 to 4). Take a look at these sample figures and the guidelines that follow. See if you can spot the mistakes and compare your answers with the answers provided at the end of this editorial.
FIGURES AND GRAPHICAL STANDARDS
Articles should contain only necessary figures, generally, no more than a combined total of six figures or tables. Figures should be concise and should be prepared with consideration for the page layout (i.e., two-column text, 8.5 × 11-inch page dimensions).
Figures should be the width of one column (3.5 inches wide) or the width of a page (7.5 inches wide) with a maximum height of 9.5 inches. Figures may be 5 inches wide only if that is required to clearly communicate essential information contained in the figure (i.e., if the figure is too crowded at 3.5 inches, but too spread out at 7.5 inches).
File Formats and Resolution
Both vector and raster graphic file formats are accepted. Graphics should be created, saved, and submitted in the following editable formats: PowerPoint (.pptx), Tagged Image File Format (TIFF), Adobe (PDF), or Encapsulated PostScript (EPS). Line art must have a resolution of at least 600 dpi at final width. Clinical images such as photographs, radiographs, optical coherence tomography, or computed tomography scans must have a resolution of at least 350 dpi at final width.
- Do not place a complete box around the figure or the plot area. Plots containing two y axes should have a bounding box around the plot area but not the entire figure.
- Keys identifying grouped data (e.g., by color, line style, markers, etc.) must be included within axes of the plot area.
- Title fonts of x and y axes must be 10-point Arial bold black.
- Axis values, keys, and all other text within the plot box must be 8-point Arial regular.
- Axis lines, tick marks, and whisker bar widths must be 0.5-point lines in black or dark gray (25% brightness).
- Data points should be 8 points in size and should not include borders, lines, or shadows.
- Trend lines must be 1.5 points and should be the same color as the data series.
- Error bars, if added, should be 1.5 points and have the same color as the data series or black.
Bar plots must be solid in color with no special formatting (e.g., no texture, gradient, 3D, or shadows).
Use the OVS Journal theme color seafoam green (red [R], 0; green [G], 148; blue [B], 127) or primary colors of sufficient contrast to be easily distinguishable by the reader.
Additional colors, in order of preference, are as follows: gray, 50% (R: 146; G: 146; B: 146); red (R: 204; G: 51; B: 51); blue (R: 51; G: 102; B: 153); and gold (R: 226; G: 152; B: 45).
For multipanel figures, each panel (subpart) should be identified with a capital letter (14-point Arial bold) in the upper left corner using a high-contrast color (e.g., white on black). For plot or line drawings, place the panel's letter above the y axis title. For images, place the panel's letter in the upper-left corner.
Figure legends are a required element; figures should be understandable apart from the text. Legends must be submitted on a separate Figure Legends page immediately after the references. Each figure legend should contain the following four key elements:
- Title: a concise summary for the whole figure, including any panels. The figure title should be placed in the figure legend. Titles could be descriptive, stating the type of experiment that produced the results, or declarative, summarizing the overall result. Titles above figures are discouraged. However, if a title above the figure is necessary, it must be in Arial 10-point bold title case, centered over the plot box.
- Description of methods: a brief description of the methods that produced the data presented in the figure (e.g., groups tested, sample sizes, and testing methods). This should be limited to the information relevant and required to understand the data presented in the figure.
- Description of results: a summary statement of the results presented in the figure, if not already included in a declarative brief title.
- Explanation of symbols: explain all symbols, colors, nonstandard abbreviations, lines, scale bars, and error bars (e.g., standard deviation vs. standard error).
See the following example:
FIGURE 1. Keratocyte density of the rabbit cornea by confocal microscopy after lamellar incision. Tangential (en face) section of the anterior corneal stroma before incision. Keratocytes are more numerous in the anterior stroma, with highly reflective nuclei and visible cytoplasm revealing cell to cell junctions. Scale bar 50 μm.
Use scale markers in images for histology, electron micrographs, or other microscopy images. Indicate any stains or contrast agents used. No keys should appear on the figures, and titles should be avoided.
Text labels and markers overlying images must be 8-point Arial regular font. Use contrasting font colors so that text is visible.
The journal encourages color figures for publication that will enhance an article. Color images must be created, saved, and submitted as RGB color files. There is no charge for color figures.
Text-only figures, such as flow diagrams, must be 10-point Arial bold for the primary headings (titles) and 8-point Arial regular for secondary headings (text). Use only black text; do not use special formatting elements such as shadows or gradient for text, lines, boxes, or other figure elements.
Case Report: Clinical Timeline
Fig. 1 must be a clinical timeline indicating key events relevant to disease presentation, examination, diagnostic testing, treatments, resolution, referrals, and so on. Authors must customize the OVS clinical timeline template for their particular case. The template is available at: http://edmgr.ovid.com/ovs/accounts/OVS_Figure_1_Clinical_Timeline_template.pptx.
The final width of this figure must be 7.5 inches. Do not reformat the figure fonts or colors. Submit this as an original PowerPoint file that can be opened and edited. Contact the Managing Editor at email@example.com with any questions.
Reuse Permissions and Copyright
Authors must submit written permission from the copyright owner to use direct quotations, tables, or illustrations previously published and copyrighted. Complete source details and citation information must be included. Any permissions fees that might be required by the copyright owner are the responsibility of the authors requesting reuse and are not the responsibility of Lippincott Williams & Wilkins or the American Academy of Optometry. To secure permissions for Lippincott Williams & Wilkins published content, go to: https://shop.lww.com/permissions.
Data tables should be used to provide a compact summary of large data sets. They should permit easy visualization and comparison of summary values and stratified data (e.g., means, confidence intervals, risk ratios, odds ratios, etc., by group). Data tables are not a suitable substitute for graphical visualizations of primary outcome measures where individual data points are presented. Tabular summaries should not be the only form of data presentation.
Tables should include a title above the table, appropriate column heads, and explanatory captions that define any abbreviations used. Tables should be self-explanatory and should supplement rather than duplicate information presented in the text. Information that can be described in two or three sentences should be provided in the text and should not be presented as a table. Tables can be formatted as long or wide, and care should be taken by the authors to carefully consider the best way to present their information. Narrow tables can be formatted to fit a single text column in width (3.5 inches), whereas wide tables can be formatted to span the full-page width (7.5 inches).
Create tables using the table creating and editing feature of your word processing software (e.g., Word). Tables must be submitted as editable text files. Picture files are not acceptable. Do not use Excel or comparable spreadsheet programs.
Fig. 1 errors: box around the plot area, y axis gridlines, y axis labels are rotated 90°, axes titles are not 10-point Arial bold, axes values are not 8-point Arial regular font, and markers are not 8-point in size.
Fig. 2 errors: box around the figure area, figure is too large, y axis has too many tick marks (plot every 10), y and x axis titles are not Arial 10-point bold, y and x axis units are not Arial 8-point regular font, bars should not be filled with candy-striped texture, and bars should not have shadow formatting.
Fig. 3 errors: box around the figure and plot areas for each panel, panels should not have title above each plot box, figure keys are placed in a different location in each panel, panel labels (A, B, C, and D) should be located above (not below) the y axis titles, y axis has too few tick marks (plot every 50), y axis labels are rotated 90°, y and x axis titles are not Arial 10-point bold, and y and x axis units are not Arial 8-point regular font.
Fig. 4 errors: none.