We use an objective statistical test to assign colour values, and apply the red and green colours only if the difference from the performance indicator's target is big enough to be statistically significant, and not just random variation.
We use standard statistical techniques to construct 95% confidence limits around our performance and apply colours as follows:
Green: The target falls outside the confidence interval on the side that indicates a positive performance.
Yellow: The target is within the confidence interval.
Red: The target falls outside the confidence interval on the side that indicates a negative performance.
Blank/White: This means one of two things:
- There is not enough data available to calculate the confidence interval.
- The colour key is not applicable to that piece of data.