Catalog

Guideline Catalog

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

781 records

Page 10 of 33

  1. Flip axis direction in parallel coordinates to depict correlation as negative when possible

    For association judgments, use axis direction and axis order on parallel-coordinate plots to improve fidelity and mitigate weaker discrimination of positively sloped relationships for people comparing correlation strength.

    • purpose:refine
    • basis:empirical
    • task:relate
    • chart:parallel
    • quality:fidelity
    • +2

  2. Flip parallel-coordinate axes to depict correlation as a negative pattern

    For relate tasks, use axis flipping or axis rearrangement on parallel-coordinate displays to depict a relationship as negative correlation to improve fidelity and mitigate less precise positive-correlation judgments for readers distinguishing nearby association strengths.

    • purpose:refine
    • basis:empirical
    • task:relate
    • chart:parallel
    • quality:fidelity:use
    • +2

  3. Format displayed data text as human-readable values

    For accessibility review of chart text, use human-readable value formatting on labels, axes, annotations, tables, and legends to improve accessibility and mitigate hard-to-parse numeric strings and unfamiliar terms for screen-reader users and readers with cognitive accessibility needs.

    • purpose:refine
    • basis:accessibility
    • quality:accessibility
    • lever:text-annotation
    • needs:screen-reader
    • +3

  4. Gray out less important categories before coloring key data

    For overview reading, use selective color emphasis with gray background categories on multi-category charts to improve focus and mitigate all-at-once emphasis for readers who need to spot the main takeaway first.

    • purpose:refine
    • basis:heuristic
    • quality:insight
    • lever:encoding
    • reading-mode:overview
    • +2

  5. Group affected icons together when an icon array must be read exactly

    For retrieve tasks in part-whole risk displays, use grouped affected icons on icon arrays to improve fidelity and mitigate slow or inaccurate counting for exact readers.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • quality:fidelity
    • lever:layout-structure
    • +2

  6. Group event icons when readers must compare risk magnitude

    For compare tasks on icon-array risk graphics, use grouped event positions on quantitative part-whole displays to improve magnitude reading and mitigate counting difficulty for mixed-numeracy audiences.

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

  7. Group many tiny areas into an 'others' category

    For dense area charts over ordered time, use grouping in the encoded series to improve readability and mitigate too many labels for readers scanning the chart.

    • purpose:refine
    • basis:heuristic
    • time:ordered-time
    • chart:area
    • quality:readability:use
    • +3

  8. Group same-category marks when order is free

    For target-search tasks, use spatial grouping of same-category marks on reorderable categorical mark views to improve search speed and mitigate slow scanning in randomly interleaved layouts for readers locating a single target.

    • purpose:refine
    • basis:empirical
    • data:categorical
    • quality:readability
    • lever:layout-structure
    • +1

  9. Group semantically related words into distinct visual zones

    For time-constrained topic-understanding tasks, prefer semantic grouping on word-cloud layouts to improve insight and mitigate category mixing for viewers using text summaries for analytic reading.

    • purpose:refine
    • basis:empirical
    • chart:word-cloud
    • data:text
    • quality:insight
    • +2

  10. Group small pie slices into an others category

    For pie charts with many small slices, use grouped slices on the chart to improve readability and mitigate clutter from tiny wedges and many labels for readers.

    • purpose:refine
    • basis:heuristic
    • chart:pie-donut
    • lever:encoding
    • group-cardinality:many
    • +2

  11. Group tiny parts into an other segment

    For dense part-to-whole stacked column charts, use category grouping for tiny segments to improve readability and mitigate label clutter for readers scanning the important parts.

    • purpose:refine
    • basis:heuristic
    • chart:bar
    • operator:part-whole
    • density:dense
    • +3

  12. Highlight only the categories that support the main message

    For explanatory multi-category charts, use color emphasis on one or a few categories to maximize insight and mitigate equal visual weight across categories for readers following a stated message.

    • purpose:refine
    • basis:heuristic
    • lever:encoding
    • channel:color-hue:use
    • group-cardinality:many
    • +3

  13. Identify anomalies separately from extreme values

    For record-list exception analysis of quantitative data, use interaction support for anomaly identification in an information visualization system to improve insight and address the mistake of treating anomalies as maxima or minima for analysts.

    • purpose:refine
    • basis:empirical
    • scope:record-list
    • data:quantitative
    • shape:outlier-rich
    • +3

  14. Include a centered Gaussian baseline

    For grouped-result comparisons of fixation-prediction models on natural image or video datasets, use a centered Gaussian baseline in the benchmark to improve trust and mitigate mistaking dataset center bias for model quality for analysts.

    • purpose:refine
    • basis:empirical
    • task:compare
    • scope:grouped-result
    • quality:trust:use
    • +2

  15. Include fixed background and foreground colors when optimizing a categorical palette

    For categorical encoding in repeated-use views, use palette optimization that includes fixed background or foreground colors on the target chart or map context to improve readability and mitigate category colors blending into surrounding display colors for analysts and designers.

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

  16. Include only parts of the whole in a stacked column chart

    For part-to-whole composition in stacked column charts, use only component parts in the stack to improve fidelity and mitigate double-counting or incomplete totals for readers interpreting the whole.

    • purpose:refine
    • basis:heuristic
    • chart:bar
    • operator:part-whole
    • quality:fidelity
    • +1

  17. Increase bubble radius when BubbleView viewing time is short

    For short-time crowdsourced importance measurement, use a larger bubble radius on dense static visuals to improve fidelity and mitigate undersampling caused by tiny focal windows in remote BubbleView studies.

    • purpose:refine
    • basis:empirical
    • time:time-interval
    • density:dense
    • quality:fidelity
    • +1

  18. Increase color contrast for small or separated marks

    For comparison in charts with small or widely separated marks, use stronger hue or lightness contrast on the colored marks to improve readability and mitigate hard-to-distinguish comparisons.

    • purpose:refine
    • basis:heuristic
    • task:compare
    • quality:readability
    • lever:encoding
    • +2

  19. Increase contrast for category-colored annotation text

    For annotations that reuse bright category colors, use higher-contrast text styling on colored words to improve readability and mitigate illegible thin letterforms for readers reading colored annotation text.

    • purpose:refine
    • basis:heuristic
    • quality:readability:use
    • lever:text-annotation
    • component:annotation:use
    • +2

  20. Increase hue spacing for small point marks

    For color-difference reading in static charts, use wider hue spacing on scatter plots with small point marks to improve fidelity and mitigate missed color differences for viewers reading web-based visualizations.

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

  21. Increase mark size until search speed levels off

    For target-search tasks in dense mark-based views, use larger mark sizes on point- or square-mark displays to improve search speed and mitigate slow localization caused by tiny marks for readers scanning for one target.

    • purpose:refine
    • basis:empirical
    • quality:readability
    • lever:encoding
    • reading-mode:lookup
    • +1

  22. Increase perceptual contrast on task-relevant features for identification tasks

    For lookup tasks, use stronger perceptual separation on the features that define the target subset in feature-coded views to improve fidelity and mitigate missed or slow target detection for viewers scanning a display.

    • purpose:refine
    • basis:empirical
    • quality:fidelity:use
    • lever:encoding
    • reading-mode:lookup

  23. Increase separation between adjacent steps in a color gradient

    For reading stepped magnitude with color, use clearly separated adjacent steps on continuous color scales to improve readability and mitigate subtle value differences for readers comparing neighboring values.

    • purpose:refine
    • basis:heuristic
    • data:quantitative
    • quality:readability
    • lever:encoding
    • +1

  24. Increase target-defining contrast for known-value identification

    For known-value target lookup, use strong task-relevant feature contrast on feature-coded displays to improve fidelity and mitigate distractor interference for readers doing lookup.

    • purpose:refine
    • basis:empirical
    • task:retrieve
    • quality:fidelity
    • lever:encoding
    • +1