Data Relationships and Chart Types

Every dataset tells one of six stories: change over time, comparison between categories, part-to-whole composition, distribution of values, correlation between variables, or geographic patterns. Each story has chart types that communicate it clearly and chart types that obscure it. A line chart communicates change over time with 94% comprehension accuracy. A pie chart communicates the same data with 62% accuracy. The choice is not aesthetic. It is functional. The chart type determines whether the audience understands the data or misinterprets it. That gap — between comprehension and confusion — is the entire purpose of chart selection.

1. Step 1: Identify the Data Relationship What story does this data tell? Change over time, comparison, composition, distribution, correlation, or geography. The answer is not always obvious — the same dataset can tell multiple stories. Choose the one that serves the decision the audience needs to make.
2. Step 2: Match to Chart Type Change over time: line chart or area chart. Comparison: bar chart. Composition: stacked bar or treemap. Distribution: histogram or box plot. Correlation: scatter plot. Geography: choropleth map. This is the starting point. Refinement follows.
3. Step 3: Refine for Context How many data points? Two series need a simple line chart. Twenty series need small multiples. How precise does it need to be? A bar chart with values labeled enables precise comparison. A treemap shows proportions at a glance but loses precision. Match the chart refinement to the audience need.