Scale

A scale is the rule that translates data values into distances along an axis, fixing the axis's range, its interval spacing, and whether equal steps represent equal amounts or equal multiples.

Scale is what makes a position on a chart mean a specific number. When the scale says "0 at the bottom, 100 at the top, with a tick every 20," then a point halfway up the axis represents 50. Change the scale and the same data can look dramatically different, which is why scale is one of the most consequential — and most easily abused — choices in a chart.

Linear and logarithmic scales

A linear scale spaces values evenly: every equal step covers the same amount, so 0–20 takes exactly as much room as 80–100. This is the default and the one most readers assume. A logarithmic scale instead spaces values by multiples — each step is ten times (or some other factor) the last, so 1, 10, 100 and 1,000 sit at equal intervals. Logarithmic scales are useful when data spans several orders of magnitude, but they must be labelled clearly because they flatten large differences.

A concrete example

Suppose a bar chart's tallest value is 90. A scale running 0 to 100 makes the bars read as honest proportions of one another. But if the scale is set to run 80 to 100, a value of 90 fills half the axis while a value of 85 nearly vanishes — the differences look enormous even though they are small. Choosing where the scale starts and stops therefore shapes the story the chart tells, before a single number is read.

Related terms

A scale always belongs to an axis — typically the numeric y-axis, though an x-axis of measurements has one too. The tick marks a scale defines are often extended across the plot as gridlines. For how scale choices affect comparison, see the bar chart guide.