Catalog

Guideline Catalog

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

781 records

Page 11 of 33

  1. Increase track count when tracks alone must convey uncertainty

    For risk judgments from uncertain geospatial paths over ordered time, use more than 15 visible trajectories on unannotated track displays to improve fidelity and mitigate underreading of spatial spread for viewers with low domain knowledge.

    • purpose:refine
    • basis:empirical
    • time:ordered-time
    • data:geospatial
    • lever:layout-structure
    • +3

  2. Insert a gap between compared segments in a single stacked bar

    For exact part comparison tasks, use visible separation on single stacked bars to improve fidelity and mitigate part-to-whole bias for viewers making quick visual estimates.

    • purpose:refine
    • basis:empirical
    • task:compare
    • chart:bar
    • quality:fidelity:use
    • +3

  3. Inspect an importance map after each layout revision

    For iterative revision of a single-result graphic design, use importance-map feedback on a single-view layout to improve visual hierarchy and mitigate unintended emphasis shifts for designers refining element order.

    • purpose:refine
    • basis:empirical
    • task:compose
    • structure:single-view
    • quality:readability
    • +3

  4. Integrate very high ratio values with the full 100% range

    For exact reading of high part-to-whole values, use integrated full-range encoding on bar charts to improve fidelity and mitigate underestimation of large values for readers judging values near the top of a 0–100% scale.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • chart:bar
    • quality:fidelity:use
    • +3

  5. Iterate visualization designs with multiple expert reviews

    For visualization development across design stages, use iterative expert review on evolving visualization prototypes to improve fidelity and mitigate chart, interpretation, and usability mistakes for mixed expert teams.

    • purpose:refine
    • basis:rhetorical
    • quality:fidelity
    • communication:workflow
    • audience:reviewer
    • +1

  6. Keep a consistent visual platform across consecutive frames

    For step-by-step narrative reading, use a consistent visual platform on slideshow or tabbed views to improve readability and mitigate viewer reorientation for readers following transitions.

    • purpose:refine
    • basis:empirical
    • structure:multi-view
    • quality:readability:use
    • lever:layout-structure
    • +1

  7. Keep a standard scatterplot preset for correlation estimation

    For relate tasks, avoid switching among tested scatterplot presets on quantitative scatterplots to prevent unsupported expectations of fidelity gains and address unnecessary preset churn in preset-selection workflows.

    • purpose:refine
    • basis:empirical
    • task:relate
    • chart:scatter
    • data:quantitative
    • +2

  8. Keep bar aspect ratios near square for memory-based value reading

    For exact recall of bar values across separated views, prefer near-square mark aspect ratios on bar charts to improve recall fidelity and mitigate overestimation of wide bars and underestimation of tall bars for viewers making memory-based readings.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • chart:bar
    • quality:fidelity
    • +2

  9. Keep bar marks close to square when readers must recall single values

    For retrieve-value tasks involving recalled single values, prefer square bar aspect ratios on bar charts to improve fidelity and mitigate systematic overestimation and underestimation for viewers comparing values across separate views.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • scope:single-result
    • chart:bar
    • +3

  10. Keep categorical palette lightness in a midrange when colors must work on black and white backgrounds

    For categorical comparison on visualizations shown against black and white backgrounds, use midrange color lightness on categorical encodings to improve readability and address disappearing or washed-out colors for viewers distinguishing categories.

    • purpose:refine
    • basis:empirical
    • data:categorical
    • quality:readability
    • lever:encoding
    • +3

  11. Keep categorical palettes to seven colors or fewer

    For category comparison, avoid large color sets on charts that rely on categorical color encoding to improve readability and mitigate repeated legend checking for readers.

    • purpose:refine
    • basis:heuristic
    • data:categorical
    • quality:readability
    • lever:encoding
    • +3

  12. Keep chart text at or above 9 pt / 12 px

    For chart reading, use text sizing of at least 9 pt / 12 px on chart text to improve accessibility and mitigate unreadable small-text failures for readers with low vision.

    • purpose:refine
    • basis:accessibility
    • quality:accessibility
    • lever:text-annotation
    • needs:low-vision

  13. Keep chart text readable instead of shrinking it to fit

    For chart text in web visualizations, use readable font settings and avoid shrinking or narrowing type to fit to improve readability and mitigate uncomfortable reading in tight layouts for readers viewing titles, labels, and annotations.

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

  14. Keep compared bars adjacent in simple bar charts

    For exact comparison tasks, prefer adjacent placement of compared bars on simple bar charts to improve fidelity and mitigate separation errors for viewers making quick visual estimates.

    • purpose:refine
    • basis:empirical
    • task:compare
    • chart:bar
    • quality:fidelity:use
    • +3

  15. Keep continuous colormaps invertible for legend lookup

    For legend-based lookup across ordered values, avoid repeated or flat color regions on continuous color scales to prevent ambiguous value inversion and address non-unique color-to-value mappings for viewers matching colors back to the legend.

    • purpose:refine
    • basis:empirical
    • quality:fidelity:use
    • lever:encoding
    • polish:palette
    • +1

  16. Keep denominator size consistent across equivalent icon arrays

    For comparison of equivalent medical risks in non-temporal icon-array displays, prefer a fixed denominator size across otherwise matched views to improve fidelity and mitigate denominator-size bias for audiences comparing treatment effects.

    • purpose:refine
    • basis:empirical
    • task:compare
    • quality:fidelity:use
    • lever:encoding

  17. Keep denominators consistent in numerical risk comparisons

    For compare tasks on grouped-result numerical risk displays, use consistent denominators across groups to improve fidelity and mitigate denominator neglect for audiences with low numeracy.

    • purpose:refine
    • basis:empirical
    • task:compare
    • scope:grouped-result
    • quality:fidelity
    • +3

  18. Keep each party mapped to the same color across election graphics

    For grouped election-result reporting across related charts and maps, use a consistent party-color mapping to improve readability and mitigate reader confusion from color reassignment for readers following election coverage.

    • purpose:refine
    • basis:heuristic
    • scope:grouped-result
    • quality:readability
    • lever:encoding
    • +3

  19. Keep gridlines at least 8 pixels apart

    For exact value comparison in static displays, use gridline spacing of at least 8 pixels on bar and line charts to improve fidelity and mitigate tracing errors from overly dense reference lines for web viewers.

    • purpose:refine
    • basis:empirical
    • task:compare
    • chart:bar
    • chart:line
    • +3

  20. Keep hypothetical outcome plot playback slow enough to perceive individual samples

    For trend judgment in ordered-time displays, use a cognitively trackable animation rate on hypothetical outcome plots to improve fidelity and mitigate loss of sensitivity from over-fast playback for novice public audiences.

    • purpose:refine
    • basis:empirical
    • task:trend
    • time:ordered-time
    • quality:fidelity
    • +4

  21. Keep increasing values higher on the y-axis

    For trend interpretation over ordered time, avoid inverted y-axis order on axis-based trend charts to prevent message reversal and mitigate false improvement or decline readings for chart readers.

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

  22. Keep margin-of-error and comparison values on the chart instead of moving them into text

    For inferential judgments of mean and error, avoid text annotation on bar charts for comparison values or margins of error to prevent inaccuracy and address failed bias mitigation for viewers without deep statistical training.

    • purpose:refine
    • basis:empirical
    • task:compare
    • chart:bar
    • quality:fidelity
    • +2

  23. Keep meaningful chart elements visually separable

    For visual interpretation of charts where adjacent elements touch or overlap, use visible separation and unobscured placement on primary marks and text to improve accessibility and mitigate merged marks, hidden labels, and obscured elements for viewers who must visually distinguish chart content.

    • purpose:refine
    • basis:accessibility
    • quality:accessibility
    • lever:layout-structure
    • polish:spacing

  24. Keep multiple temporal states visible for analytical comparison

    For analytical tasks on ordered-time data, avoid substitutive animation in temporal views to improve fidelity and mitigate recall burden for users comparing changing states.

    • purpose:refine
    • basis:empirical
    • task:trend
    • time:ordered-time
    • temporal-pattern:dynamic
    • +2