Catalog

Guideline Catalog

Browse visualization guideline records with sections, labels, and references.

781 records

Page 30 of 33

  1. Use line charts instead of pie charts for correlation judgments

    For correlation judgments in small static two-dimensional displays, use a line chart instead of a pie chart on tabular data to improve fidelity and speed and mitigate weak association reading for readers judging whether two attributes move together.

    • purpose:select
    • basis:empirical
    • task:relate
    • chart:line:use
    • chart:pie-donut:avoid
    • +3

  2. Use line charts to convey continuous trend

    For trend reading on ordered quantitative data, prefer a line chart on displays of sequential observations to improve readability and mitigate discrete-comparison interpretations for readers.

    • purpose:select
    • basis:empirical
    • task:trend
    • chart:line:use
    • chart:bar:avoid
    • +2

  3. Use line charts to foreground x–y interactions

    For open-ended interpretation of grouped three-variable data, use a line chart instead of a bar chart on multivariate quantitative graphs to improve insight into x–y interactions and mitigate unintended focus on legend-variable contrasts for viewers identifying the main point.

    • purpose:select
    • basis:empirical
    • task:relate
    • chart:line:use
    • chart:bar:avoid
    • +3

  4. Use lining tabular numerals for chart values

    For exact reading of values in charts and tables, use lining tabular numerals on labels, axis ticks, tooltips, and table cells to improve readability and mitigate misalignment from varying digit heights and widths for readers comparing numbers.

    • purpose:refine
    • basis:heuristic
    • data:quantitative
    • quality:readability
    • lever:text-annotation
    • +1

  5. Use low-density quantile dotplots when viewers must estimate probabilities precisely

    For exact probability-interval estimation in static timeline views of continuous predictive uncertainty, use low-density discrete-outcome encoding on timeline rows to improve fidelity and mitigate imprecise interval reading for novice mobile users.

    • purpose:refine
    • basis:empirical
    • chart:timeline
    • operator:uncertainty
    • reading-mode:exact
    • +4

  6. Use maps or geo-referencing when location matters

    For explanation or comparison where geographic scope is part of the message, use maps or geo-referencing on place-based data displays to improve spatial understanding and address missing place context for readers relating data to known places.

    • purpose:refine
    • basis:rhetorical
    • chart:map:use
    • data:geospatial
    • reading-mode:overview
    • +2

  7. Use mortality framing when comparing treatment curves across different time spans

    For ordered-time treatment comparisons, use mortality framing on line charts to improve fidelity and mitigate temporal inconsistency bias when readers compare effectiveness across displays with different year spans.

    • purpose:refine
    • basis:empirical
    • task:compare
    • time:ordered-time
    • chart:line
    • +3

  8. Use motion on the key changing element

    For ordered-time explanatory visuals, use motion on the key changing element in a dynamic display to improve readability and mitigate uncontrolled viewing order for viewers following an unfamiliar process.

    • purpose:refine
    • basis:empirical
    • time:ordered-time
    • temporal-pattern:dynamic
    • quality:readability:use
    • +1

  9. Use multiple distinct colors when memorability is the goal

    For short-exposure recall of single static visualizations, use multiple distinct colors on the chart to improve memorability and mitigate bland, easily confusable displays for viewers scanning many graphics.

    • purpose:refine
    • basis:empirical
    • quality:insight:use
    • lever:encoding
    • aesthetic:color:use
    • +1

  10. Use multiple plausible paths instead of a single widening cone to show track uncertainty

    For intuitive impact judgments across ordered future time, use a multiple-instance uncertainty encoding on geospatial forecast maps to improve fidelity and mitigate misreadings of forecast uncertainty as storm growth for viewers with low domain knowledge.

    • purpose:refine
    • basis:empirical
    • time:ordered-time
    • data:geospatial
    • quality:fidelity
    • +3

  11. Use multiplexed numerals when bolding table values

    For exact reading of highlighted values in tables, use multiplexed numerals on bold and regular table text to improve readability and mitigate alignment shifts when emphasized numbers change weight for readers scanning columns.

    • purpose:refine
    • basis:heuristic
    • chart:table
    • data:quantitative
    • quality:readability
    • +3

  12. Use natural frequencies instead of percentages for diagnostic and screening risks

    For interpreting single-risk results in diagnostic or screening contexts, prefer natural-frequency wording on quantitative risk labels and captions to improve fidelity and mitigate incorrect risk estimates for health professionals and consumers.

    • purpose:refine
    • basis:empirical
    • scope:single-result
    • data:quantitative
    • quality:fidelity:use
    • +3

  13. Use nested containing frames when the treemap must show parent levels

    For overview of hierarchical treemaps, use nested containing frames on a treemap to improve readability and address hidden containment when users need parent levels or container names visible.

    • purpose:refine
    • basis:empirical
    • chart:treemap
    • data:hierarchical
    • quality:readability
    • +2

  14. Use no more than four shapes in grouped scatterplots

    For grouped scatterplots, prefer shape encoding on point marks to improve accessibility and mitigate color-only grouping, and avoid large shape sets that create visual clutter for readers with color-vision deficiency.

    • purpose:refine
    • basis:heuristic
    • chart:scatter
    • quality:accessibility
    • lever:encoding
    • +3

  15. Use non-contiguous cartograms for region shape recognition

    For lookup tasks on geospatial cartograms, prefer a shape-preserving cartogram type on distorted map views to improve fidelity and mitigate region-shape identification errors for readers matching regions back to the original map.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • chart:map
    • data:geospatial
    • +3

  16. Use normal-width fonts for chart text

    For labels and numbers under space constraints in charts, prefer normal-width typefaces on chart text to improve readability and mitigate cramped text from condensed fonts for readers scanning values.

    • purpose:refine
    • basis:heuristic
    • quality:readability
    • lever:text-annotation
    • aesthetic:style:use

  17. Use one accent color and same-hue shades in a pie chart

    For emphasizing one important share in a pie chart, use a focused slice-color palette to improve readability and mitigate distracting rainbow colors for readers.

    • purpose:refine
    • basis:heuristic
    • chart:pie-donut
    • lever:encoding
    • quality:readability
    • +3

  18. Use one hue per parent category and shades for its subcategories

    For grouped categorical displays with nested categories, use hue for the parent groups and lightness variation within each hue for subcategories to improve readability and address overly fragmented category palettes for readers separating group levels.

    • purpose:refine
    • basis:heuristic
    • data:hierarchical
    • quality:readability
    • lever:encoding
    • +2

  19. Use oriented glyphs instead of oriented textures for average-direction summaries

    For overview summary tasks, prefer oriented glyphs on directional displays to improve fidelity and mitigate imprecise average-direction judgments for viewers summarizing local regions.

    • purpose:refine
    • basis:empirical
    • quality:fidelity:use
    • lever:encoding
    • channel:orientation:use
    • +2

  20. Use overlaid bars for static biggest-mover comparison

    For comparison of two quantitative series in a biggest-mover task when the view must stay static, use an overlaid single-view layout on bar charts to improve fidelity and mitigate cross-panel comparison errors for viewers inspecting one pair at a time.

    • purpose:select
    • basis:empirical
    • task:compare
    • chart:bar
    • structure:single-view:use
    • +3

  21. Use pair-preference-weighted hue palettes when palette liking is the main goal

    For categorical palette design when palette liking is the main goal, prefer a pair-preference-weighted hue palette on nominal color encodings to maximize aesthetics and address harsh category combinations for readers judging the palette as a whole.

    • purpose:refine
    • basis:empirical
    • data:categorical
    • quality:aesthetics:use
    • lever:encoding
    • +2

  22. Use personal-reference framing to connect data to everyday experience

    For public-facing explanatory visualizations intended to motivate or connect, use personal-reference text annotation on chart framing to improve insight and address detached, impersonal presentation for general-public viewers.

    • purpose:refine
    • basis:rhetorical
    • quality:insight
    • lever:text-annotation
    • communication:resonance
    • +1

  23. Use pictograms only when they reinforce the chart's message

    For recall-oriented explanation of a single visualization, use pictorial shapes on the chart only when they reinforce the data or message to improve insight and address weak recognition for readers who need a visual hook.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • quality:insight
    • lever:encoding
    • +2

  24. Use pictograph frequency displays for comparative risk and benefit counts

    For comparative decision communication of quantitative risks and benefits, use pictograph frequency encoding on risk/benefit displays to improve gist and verbatim understanding and mitigate numerical misinterpretation for lay audiences with mixed numeracy.

    • purpose:refine
    • basis:empirical
    • task:compare
    • data:quantitative
    • quality:fidelity
    • +3