Lab 07 / Communication

Points are not percent

Is a move from 4% to 6% a 2% or 50% increase?

4 min synthetic data fixed seed

Set a baseline and a new rate, then compare percentage points, relative change, and absolute counts side by side. Change one control at a time, then open the values under the chart before interpreting the picture.

Interactive synthetic lab
runs in this browser

View the values behind this chart

Loading the deterministic simulation...

SEED

What this display answers

Percentage points subtract two rates directly. Relative percent change divides that difference by the baseline. Both can be correct, but they answer different questions.

Absolute counts add the denominator that either percentage can conceal. A two-point move has different practical scale in a population of 1,000 than in 100,000.

What to notice

  1. Start at 4% and 6%: the result is +2 percentage points and +50% relative change.
  2. Set the baseline to zero. The point change still exists, but relative change is undefined.
  3. Change only the reference count. Rates stay fixed while the count difference changes.

The model behind it

Rates are entered on a 0-100 percent scale and applied to one fixed synthetic reference count.

The controls use whole percentages and reference populations divisible by 100, so displayed expected counts are whole numbers. They are model outputs, not observed people.

Where the result stops

Relative changes around a very small baseline can sound dramatic while representing a small absolute difference.

Always state the denominator, time window, and whether a reported percent means rate, share, or relative change.