Charts and Graphs for Research Posters: A Decision Guide

Choose from the comparison task

Use bars or dots for categories, lines for ordered change, scatterplots for relationships, and distributions when variation matters. Avoid choosing a form because it looks modern. The chart should answer a sentence-level question stated in the heading.

• Use honest baselines.
• Order categories meaningfully.
• Reserve pie charts for simple shares.

Remove legend hunting

Direct labels, selective annotations, and clear units reduce memory work. Remove duplicate gridlines, borders, shading, three-dimensional effects, and decorative icons. Highlight one focal series while keeping context legible, and ensure every encoding works without colour.

• Put units near values.
• Explain error bars.
• Avoid tiny manuscript panels.

Audit the claim after redesign

Changing chart type or scale can change the impression even when numbers remain constant. Check baseline, missing data, uncertainty, and absolute versus relative effects. Let a subject reviewer compare the redesigned figure with the approved analysis.

• Retain the source figure.
• Verify final annotations.
• Export vectors when possible.

A practical example

For five treatment groups with intervals, a sorted dot-and-whisker plot often shows ranking and uncertainty more honestly than five filled bars. Direct labels can remove the legend.

Rebuild dense tables for poster distance

Custom design can translate manuscript graphics into clear, editable poster figures while keeping units, intervals, labels, and source values reviewable.

Sources and further reading

• Creating clear and informative image-based figures for scientific publications — PubMed Central
• Choosing color palettes for scientific figures — PubMed Central
• Preparing and Presenting Effective Research Posters — PubMed Central