Scatter Clusters Example (Old Faithful Geyser Eruptions)

Live Preview

Scenario

Explore how Old Faithful eruption duration relates to the waiting time before the next eruption, and identify the visible clusters in those observations.

Data Source And Refresh

• Source: Old Faithful geyser eruptions sample dataset used in ParaCharts demos.
• Source URL: https://raw.githubusercontent.com/vincentarelbundock/Rdatasets/master/csv/datasets/faithful.csv (reference dataset).
• Structure: One scatter series with eruption duration on the x-axis and waiting time on the y-axis.
• Refresh cadence: Static reference dataset.

When To Use

Use a scatter chart when the question is about clustering, separation, or outliers between two numeric variables.

Manifest

<para-chart manifest="data/manifests/old-faithful-scatter.json"></para-chart>

Full manifest (JSON):

• https://raw.githubusercontent.com/mgifford/ParaCharts/main/docs/data/manifests/old-faithful-scatter.json

Interpretation Prompts

1. Do the points separate into distinct eruption patterns, or form a single continuous cloud?
2. Which cluster appears to contain the longest waiting times?
3. Are there points that sit clearly outside the main groups?