Quantifying Uncertainty · Lesson 6 of 6
Communicating Uncertainty Honestly
~12 min
The Concept
Your final task for this mission: brief the town council on next month's streamflow forecast, including uncertainty, in language that won't be misunderstood as either false precision or unhelpful vagueness.
Communicating uncertainty well means resisting two opposite failures: hiding the uncertainty to sound more authoritative than the evidence supports, and burying the answer in so many caveats that nobody can act on it. Good uncertainty communication states a clear central estimate, a clear range, and one plain sentence about what that range means for the decision at hand.
The Analogy
A good uncertainty statement is like a doctor telling you 'recovery usually takes 4 to 6 weeks, most likely around 5,' instead of either promising an exact date they cannot really know, or listing every conceivable complication until you have no idea what to plan for.
Why Real Researchers Care
Public-facing scientific communication, hurricane forecast cones, epidemic case projections, economic forecasts, succeeds or fails largely on how well it balances these two failure modes. Mission 11 will go deeper into scientific communication generally; this lesson is the uncertainty-specific piece of that skill.
Quick Check
Q1. What are the two opposite failure modes in communicating uncertainty?
Your Goal
Write the exact two sentences you'd say to the town council about next month's streamflow, including your uncertainty, in language a non-statistician would find both clear and honest.
Hint: Try: 'We expect streamflow near X. Based on similar past months, the true value will likely fall between Y and Z.'
Teach It Back
Explain, in your own words, why 'the model says 14.2 m³/s' is a worse thing to tell a policymaker than 'we expect roughly 12-16 m³/s, with 14.2 as our best estimate'.