An uncertainty interval (UI), or plausible range, is a range of values that likely contains the true value of something unknown. It is used when data is limited or unavailable. Instead of relying only on statistics, a UI is built using expert judgment, logical reasoning, and existing knowledge. For example, if we estimate disease prevalence in a species but have no direct data, we start with a wide range and narrow it by eliminating impossible values. The UI helps express uncertainty clearly while allowing for adjustments as more information becomes available.

A 90% UI is an estimate that provides an upper bound and lower bound to a range of values that should – in your opinion – contain the actual value, 90% of the time. You can start with an absurdly wide range, then work to reduce the breadth by eliminating values you know to be extremely unlikely. For example, if asked to provide a 90% UI for the prevalence of type 2 diabetes in adults over 40 years old, you could start with 1-90%. You can quickly rule out extremely high values – it’s impossible that 90% or even 70% of adults over 40 have diabetes, as this would be a widely known crisis affecting most families. You can also rule out extremely low values like 1%, as diabetes is known to be a common condition. Even with just general knowledge from news coverage and personal experience, you might reasonably narrow to an upper bound of 35% and a lower bound of 10% (i.e., 10% to 35%). This wide range still reflects significant uncertainty, but eliminates clearly impossible values. Examining the literature could help make this interval narrower

When making your estimates, you can use a similar approach: start with the widest plausible range, then systematically narrow it by eliminating values that your expertise tells you are extremely unlikely. Draw on both published research and your field experience. If you find it difficult to narrow the range substantially, that’s perfectly acceptable – it’s better to provide a wider interval that you’re confident contains the true value than to force artificial precision.

If you wish, you can familiarize yourself with making 90% uncertainty interval estimates using an online calibration tool. You can visit this website, select ‘confidence interval’ and set it to 90%.