figure presentation issues
Broken axes, missing error bars, color-only encoding.
What it is
A figure can be technically correct and still mislead: broken y-axes that exaggerate small differences, plotted means without error bars hiding variance, log-scale axes labelled as linear, color-only encoding that fails for colourblind readers, fonts too small to read at journal column width. These aren't fraud; they're presentation choices that change what the reader perceives.
Why a reviewer cares
Reviewers notice when the visual narrative is louder than the data. They'll ask for redraws, raw values, or alternative encodings. Persistent presentation problems suggest carelessness, or worse, an attempt to nudge readers past a weak result.
How to fix it
Use continuous y-axes unless there's a compelling reason for a break. Show error bars on plotted means, with a caption that names what they are (SD, SEM, 95% CI). Don't rely on colour alone. Pair it with shape or pattern. Test legibility at journal column width before submission.
This is one of ~15 canonical methodology explainers Paper Review's red-team report links to. To get a full review of your manuscript, start a Paper Review ($9).