A laboratory-based study of understanding of uncertainty in 5-day site-specific temperature forecasts

The impact of presenting uncertainty in 5-day location-specific temperature forecasts on the decision making of non-specialists was tested in an experimental economics laboratory. Undergraduate students studying a range of disciplines were asked to select which of two criteria involving temperature would be most likely to occur based on a given 5-day forecast. If they selected a criterion that was subsequently satisfied they were given a small cash reward. It was found that students presented with uncertainty information (the 50th and 90th percentile confidence intervals) in addition to the expected temperature were more likely to select the most probable criterion. This was true irrespective of the academic subject the participants were studying.