Catalog

Guideline Catalog

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

781 records

Page 9 of 33

  1. Encode regional values with framed rectangles instead of shading

    For regional quantitative comparison, use framed-rectangle encoding on statistical maps to improve judgment fidelity and mitigate shading, area, and clustering errors for readers comparing regions.

    • purpose:refine
    • basis:empirical
    • chart:map
    • data:geospatial
    • quality:fidelity
    • +3

  2. Encode speed with color shading instead of path thickness on curved 2D trajectories

    For speed-extremum lookup on ordered-time trajectories, prefer color shading for speed on curved 2D path views to improve fidelity and mitigate curvature-driven misreading from path thickness for readers locating fastest or slowest points.

    • purpose:refine
    • basis:empirical
    • task:extreme
    • time:ordered-time
    • quality:fidelity:use
    • +3

  3. Encode the primary quantitative field with position for exact value tasks

    For exact individual-value tasks, use position encoding for the primary quantitative field on multivariate point plots to improve fidelity and mitigate misreads from less precise size or color encodings for readers performing lookup and pairwise comparison.

    • purpose:refine
    • basis:empirical
    • lever:encoding
    • channel:position:use
    • operator:lookup
    • +3

  4. Encode the primary quantitative field with size for grouped summary tasks in point plots

    For grouped summary tasks, use size encoding for the primary quantitative field on categorical point plots to improve fidelity and mitigate weaker group-level judgments from position-only encodings for readers comparing maxima or averages across categories.

    • purpose:refine
    • basis:empirical
    • lever:encoding
    • channel:area:use
    • scope:grouped-result
    • +3

  5. Encode threshold probability cumulatively instead of as density area

    For threshold uncertainty judgments at a timepoint, use cumulative position encoding on predictive distributions to improve fidelity and address area-based probability estimation for non-expert decision-makers.

    • purpose:refine
    • basis:empirical
    • quality:fidelity:use
    • lever:encoding
    • operator:uncertainty
    • +3

  6. Encode time as vertical position when 2D overlap hides movement

    For ordered-time analysis of geospatial trajectories, use vertical position for time on a 3D space-time view to improve insight and address overlapping 2D traces for readers reconstructing movement patterns.

    • purpose:refine
    • basis:empirical
    • task:trend
    • time:ordered-time
    • data:geospatial
    • +3

  7. Encode time-series magnitude with color for mean and variance judgments

    For overview summary tasks in ordered-time displays, prefer color encoding on time-series views to improve fidelity and mitigate inaccurate mean or variance estimation for viewers scanning aggregated temporal patterns.

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

  8. Encode time-series magnitude with position for extrema and range judgments

    For extreme-finding tasks in ordered-time displays, prefer position encoding on time-series views to improve fidelity and mitigate missed maxima, minima, or range judgments for viewers scanning temporal patterns.

    • purpose:refine
    • basis:empirical
    • task:extreme
    • time:ordered-time
    • data:temporal
    • +3

  9. Encode uncertain time predictions as a low-density quantile dotplot

    For quick repeated threshold decisions, use a discrete frequency encoding on uncertain time-to-event predictions to improve fidelity and address oversimplified interval summaries for non-expert users.

    • purpose:refine
    • basis:empirical
    • chart:dotplot:use
    • quality:fidelity:use
    • lever:encoding
    • +2

  10. Encode uncertainty with a symmetric opacity gradient around the mean

    For judging likelihood or comparing mean estimates with uncertainty, use a continuous opacity encoding on mean-and-error charts to improve fidelity and mitigate binary interval reading for viewers making inferential judgments.

    • purpose:refine
    • basis:empirical
    • task:compare
    • quality:fidelity
    • lever:encoding
    • +2

  11. End multi-attribute risk materials with a summary table

    For comparing many risks and benefits across options in a multi-page or extended explanation, use a summary table at the end of the material to improve readability and mitigate memory load for patient decision-makers.

    • purpose:refine
    • basis:empirical
    • task:compare
    • chart:table:use
    • structure:multi-view
    • +3

  12. Enlarge and isolate the element you want to rank highest

    For composing a single-result graphic design, use element size and placement on a single-view layout to improve visual hierarchy and mitigate weak emphasis for designers adjusting relative importance.

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

  13. Ensure color-only interactive state changes reach 3:1 contrast

    For interactive chart navigation and selection, use high-contrast color state changes on interactive elements in charts to improve accessibility and mitigate subtle hover, focus, and selection changes for low-vision users.

    • purpose:refine
    • basis:accessibility
    • quality:accessibility
    • lever:interaction-access
    • needs:low-vision
    • +1

  14. Evaluate scatterplot correlation readability with dispersion-based visual features

    For correlation comparison, prefer encoding review criteria on scatterplots to improve fidelity and address misleading association judgments for designers by inspecting dispersion-based visual features rather than Pearson correlation alone.

    • purpose:refine
    • basis:empirical
    • task:relate
    • chart:scatter
    • quality:fidelity
    • +3

  15. Explain every data-carrying color with a color key

    For charts where color encodes values or variables, use explicit color keys on the chart to improve readability and mitigate ambiguous color meaning for readers.

    • purpose:refine
    • basis:heuristic
    • quality:readability
    • lever:text-annotation
    • component:legend:use
    • +1

  16. Explain required prior knowledge on the chart

    For explanatory communication to mixed-familiarity audiences, use text annotation on charts or maps to improve readability and mitigate confusion from unexplained terms or unlabeled locations for viewers with low domain knowledge.

    • purpose:refine
    • basis:rhetorical
    • quality:readability
    • lever:text-annotation
    • communication:resonance
    • +3

  17. Explain that visible paths are a sample, not the full set

    For point-location judgments in static uncertainty forecasts, use explanatory instructions on ensemble path maps to improve fidelity and mitigate deterministic readings of individual paths for novice viewers.

    • purpose:refine
    • basis:empirical
    • chart:map
    • data:geospatial
    • quality:fidelity
    • +3

  18. Explain unfamiliar concepts in captions or surrounding text

    For explanatory reading of charts with non-obvious statistical or temporal reference concepts, use surrounding explanatory text on captions or nearby copy to improve fidelity and address confusion or distrust for viewers with low domain knowledge.

    • purpose:refine
    • basis:rhetorical
    • quality:fidelity
    • lever:text-annotation
    • communication:context
    • +1

  19. Expose a distribution summary for quantitative attributes

    For grouped-result distribution analysis of quantitative data, use interaction support for distribution summaries in an information visualization system to improve insight and address judgments about normalcy or case location within a spread for analysts.

    • purpose:refine
    • basis:empirical
    • task:distribute
    • scope:grouped-result
    • data:quantitative
    • +3

  20. Expose chart information through non-visual text alternatives

    For reading charts without visuals in static, interactive, or animated displays, use screen-reader-accessible text alternatives on chart content, annotations, and narrative elements to maximize accessibility and address information that would otherwise be communicated only visually for screen-reader and braille-reader users.

    • purpose:refine
    • basis:accessibility
    • quality:accessibility
    • lever:interaction-access
    • needs:screen-reader
    • +1

  21. Express treatment effects as absolute risk reduction

    For treatment-benefit comparison in non-temporal numerical risk displays, prefer absolute risk reduction wording on baseline-versus-treatment summaries to improve fidelity and mitigate misreading of treatment benefit for mixed-numeracy audiences.

    • purpose:refine
    • basis:empirical
    • task:compare
    • quality:fidelity:use
    • lever:text-annotation
    • +3

  22. Express uncertainty with numerical probabilities or intervals

    For communicating uncertainty in results or recommendations, use numerical uncertainty labels on annotations, legends, or captions to improve fidelity and mitigate inconsistent interpretation for readers unfamiliar with scientific verbal conventions.

    • purpose:refine
    • basis:empirical
    • quality:fidelity:use
    • lever:text-annotation
    • operator:uncertainty
    • +3

  23. Facet categories instead of coloring a single scatterplot when summary tasks face many dense groups

    For grouped summary tasks with many categories or dense overlap, prefer small-multiple faceting over a single-view category-colored scatterplot to improve fidelity and mitigate overplotting errors for readers comparing groups.

    • purpose:select
    • basis:empirical
    • structure:small-multiples:use
    • structure:single-view:avoid
    • scope:grouped-result
    • +3

  24. Filter dense link views to the strongest relationships

    For relationship analysis in dense link-based views, use a filtered subset of the strongest links to improve readability and mitigate link overload for analysts exploring many possible connections.

    • purpose:refine
    • basis:empirical
    • task:relate
    • operator:association
    • density:dense
    • +2