Calculation Average Vs Threshold Values

Entity_calculation rows with the entity_type field marked as an 'average' means that it is an average or mean of the scores/values intrinsically specified in that entity group, while the threshold is the cutoff point at which a value qualifies to be a part of that very group.

In the case of National top 10% on a more_is_better measure, an entity_calculation that is commonly used in IPRO applications and websites, the threshold value can be 95% for a measure, meaning that once a value for a hospital/geography/etc hits the 95% or above mark, it qualifies as within the top 10% threshold.  However, for the calculation for National top 10% with an entity_type of 'average' takes all of the values for the entities that have a score above the threshold value, and calculates the mean of those values.  

5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90,| 95, 100 |

For example see the above values.  In the above example, in order to fall into the Top 10%, a value must be at or above 95.  Therefore the threshold value is 95.  All the values for that measure greater than or equal to 95 (marked red) are then taken and averaged, giving us a mean of 97.5, or the Top 10% average value.

For the measures above, top 10% threshold = 95 and top 10% average = 97.5

It is uncommon for the average value to equal the threshold value, except in cases of small sample size or where the threshold and average both equal the best score possible (examples are 100% in percentage more_is_better measures like Process of Care measures, or 0% in less_is_better measures, like Mortality and Readmissions rates)