Use a map when the task is to see geographic spread

Choose a map when the job is to show where events or cases are distributed across space, and do not collapse that task into a bar chart. For example, use a map-like layout to show spread across islands or regions, and keep a bar chart only when the task is a simple count comparison by place rather than spatial spread.

A map keeps location, adjacency, and distribution visible in the display itself. A bar chart removes that spatial frame, so viewers must reason about spread without seeing where places are relative to each other.

Mechanism: Spatial arrangement supports direct reading of geographic patterns, while categorical bar positions hide the structure needed for spread analysis.

Evidence: The paper states that using a bar chart to represent the geographical spread of disease is less effective than using a map, and it also notes that the bar-chart alternative may still be better for a simple comparison task rather than a spatial one (Ola & Sedig, 2016).

• User Goal: See how a condition or event is distributed across places.
• Task: Explore spread, clustering, or geographic pattern.
• Data: Values are tied to identifiable locations or regions.
• Chart Setting: The visualization is meant to support analysis rather than only a simple categorical comparison.
• Success Criterion: Viewers can read spatial pattern directly from the display.

Break it when: The primary task is a simple comparison of counts across named places rather than understanding geographic spread. Why: The paper explicitly notes that a bar-chart view may be better suited to that simpler comparison task.

Sacrifice: You give up the very direct count-only comparison of a simple bar chart. Risk: A map adds spatial structure that may be unnecessary when geography is not part of the question. Mitigation: Keep the map for spread analysis, and use a separate simple count view only if the comparison task is also important.

Mistake: Using a bar chart to show geographic spread. Why it fails: The viewer sees counts by place name but not the spatial relationships that define spread.

Failure Sign: The viewer can compare place totals but cannot directly tell which places are near each other or how the pattern spreads across space. Quick Check: Compare a bar-chart version and a map version of the same data and ask which one lets a reviewer see spatial pattern without extra recall. Stronger Test: Ask a reviewer to describe the spread pattern directly from the chart; if they must reconstruct geography from labels alone, the bar chart is the wrong choice.

• Replace the categorical bar layout with a map-based spatial layout.
• Put the values on or within their geographic locations rather than on a categorical axis.
• Keep a separate bar chart only if you also need a simple by-place count comparison.