The first honesty check is whether the axis starts at zero, with any truncation clearly marked

What it is

The first item on the honesty checklist is the axis: does this graphic's encoding channel require a zero baseline? Does the axis start at zero? If truncated or broken, is the truncation explicitly marked? This item leads the list because axis origin is the most widespread and most easily detected of all visual distortions—it determines the reader's entire first impression of "how big the difference is."

Why it happens

The axis check's effectiveness comes from covering every combination of encoding channel and baseline: length encodings (bars, columns, areas) require zero baselines—nonzero is distortion; position encodings (dots, lines) permit nonzero origins but require annotation; radial encodings (pies, radars) are most sensitive to origin truncation (sector-area distortion amplifies nonlinearly). The check's actions are executable: read the axis origin, look for break symbols, verify that bar-height ratios match numeric ratios. A checklist uses verifiable actions ("compare numeric ratio to visual ratio") rather than principle-level reminders ("don't mislead") because principles cannot be executed while checks can—a checklist item's value lies in mapping to a concrete action with pass/fail outcomes.

Where it stops holding

The axis check cannot substitute for judging the encoding channel: the same nonzero origin is legal on a dot chart and illegal on a bar chart, so the check must confirm chart type before applying rules. The check should also cover secondary axes: a dual-axis chart's second axis is easily missed and is precisely the most commonly manipulated position. Axis-break annotation itself needs checking too—does the symbol exist, and is the drawn gap proportionate to the numeric discontinuity (drawing a tenfold data gap as a small sliver is also distortion)?

Applying it

• Check action: for each chart, record "chart type / encoding channel / axis origin / break symbol present / numeric ratio vs visual ratio"—five lines completes the audit.
• Fail length-encoded charts with nonzero origins; fail position-encoded charts with unannotated nonzero origins.
• Add this item as a required field in the chart publishing template; unchecked means unpublished.
• Verification: sample 10 recent charts, run this check, and tally failures; the failure rate is the team's true baseline on axis honesty.

Related

• Same group: U10.04.2 Check whether data range, filter conditions, and exclusions are stated · U10.04.3 Check whether title and annotation claims are supported by the graphic itself · U10.04.4 Check whether uncertainty, missing proportion, and sample size are labeled · U10.04.5 Check whether the conclusion changes under another reasonable rendering · U10.04.6 The checklist runs before publication, not after being challenged
• Nearby: U10.01.1 Truncated axes exaggerate differences · U10.01.3 A nonzero baseline is an error under length encoding
• Search terms: chart checklist · axis check · honest visualization