Breeding value estimates of bulls are calculated using data collected from Belgian Blue Breeders' Association members who participate in progeny testing.
The calculation of a breeding value estimate is done through a mathematical model. This allows the average value for a given trait to be assigned the reference value 100. When the breeding value estimate for a particular bull is less than 100, the bull scores worse than the average for that trait; if the value is greater than 100, the bull scores better than the average.
The data registration for the progeny survey started in 2019 when this studbook association made its relaunch. This means that the reference of all breeding value estimates (with an average value of 100) is currently formed by the active cows born up to 7 years back (i.e., born since 2015).
For a progeny survey to be sufficiently representative, enough progeny must be surveyed. So: the more progeny included in this survey, the more reliable the index.
Also important is the number of participating farms. Not every farmer evaluates the requested traits in the same way. Consequently, a correction must be made for a farming effect, and this is achievable if data come from many participating farms and many animals. Thus: the more farms involved in the study, the more reliable.
The association wants to be able to publish breeding values with the highest possible reliability. Therefore, for each bull, it aims for the participation of at least five different farms, each with ten offspring participating in this study. The reliability is expressed with a number from 0 (= not reliable) to 1 (= very reliable).