Choose small multiples instead of animation for temporal trend analysis

Choose a small multiples layout instead of a single animated view when analysts must find anomalies in temporal multivariate data. For example, give each item its own shared-axis panel rather than asking analysts to track many moving bubbles in one animated bubble chart.

Small multiples remove motion tracking from the task and let readers inspect each item’s path directly. That makes anomalies visible without repeated playback and reduces the overlap that hides counter-trends in one shared animated view.

Mechanism: Separate panels turn a moving tracking task into a static comparison task, so analysts can inspect paths directly instead of replaying motion to recover what happened.

Evidence: In analysis tasks, small multiples were significantly faster than animation and significantly more accurate, and participants also judged small multiples less cluttered than animation on large datasets (Robertson et al., 2008).

Notes: The paper also notes that small multiples can require serial scanning across many panels.

• User Goal: Find anomalies or counter-trends in unfamiliar data.
• Task: Analysis rather than passive presentation.
• Data: Many item-level trends changing over time in more than one visual dimension.
• Chart Setting: A scatter or bubble view is being used to show temporal movement, and no presenter is directing attention.
• Audience: Analysts exploring data on their own.
• Success Criterion: Higher answer accuracy with less time spent replaying or tracking motion.

Break it when: The primary job is a passive presentation and fast audience response matters more than accuracy. Why: Animation was the fastest condition in presentation mode and was judged more enjoyable and exciting.

Sacrifice: You give up one compact moving overview. Risk: Readers may need to scan across many panels to answer some questions. Mitigation: Use the small multiples view for analysis, then make a separate presentation view if you need a narrated walkthrough.

Mistake: Keeping all items in one animated panel and expecting anomalies to pop out during analysis. Why it fails: Analysts must replay the motion and still lose track of moving points.

Failure Sign: Analysts replay the animation several times or report losing track of points. Quick Check: Run one representative analysis task once with the animated view and once with the small multiples view; if the animated view takes longer or misses anomalies, keep the small multiples layout. Stronger Test: Verify that the answer can be found from the static panel grid without replay.

• Split each item’s temporal path into its own panel.
• Keep the axes shared across panels so paths stay comparable.
• Replace the single animated analysis view with the panel grid for exploration.
• If you still need a presentation view afterward, build a separate animated version for the walkthrough.