Using analytic confidence to improve communication and decision making

Analytic confidence is a standard created by the National Intelligence Council to standardize and apply a rigorous standard to Intelligence estimates. In the civilian world, this same standard can be applied to almost any information or judgement that’s communicated to others.

Instead of simply speaking in absolutes (yes, no) or trying to attach an arbitrary percentage of certainty to a non-numeric evaluation one can describe his confidence as high confidence, moderate confidence or low confidence. Each reflecting a level of certainty in the information, evaluation or judgement being communicated.

What do high confidence, moderate confidence and low confidence mean?

High ConfidencePossible to render a solid judgement but still carries a small risk of being wrong. Based on high quality, cohesive information
Moderate Confidence Greater possibility of being wrong but evaluation is credible. Enough information to have a make a credible evaluation but information is insufficient or uncorroborated.
Low Confidence Credible evaluation impossible Untrustworthy or very incomplete information used. Many unknowns.

So when reporting an evaluation of information, we describe our certainty as high, moderate or low. We base that description on our data. Did we have ample data? Did it all agree? Did it come from a trustworthy source? Did it come from multiple trustworthy sources?