The purpose of selecting animals is to increase the frequency of genes that control desired traits within the population. This can be done by either selecting between populations (using elite bucks from a buck breeding farm) or by selecting within a population (breeding the best does to the best bucks available).
When selection occurs within a closed population, adjustments for environmental effects must be made. These factors are not genetic and include age, birth type (single, twin or triplet), sex (males are larger) and age of dam (maiden does have smaller kids than older does). The effect of these variables can be mathematically minimized within each individual’s performance record, leaving any remaining differences attributable to genetic value. Many of the cashmere goat fiber traits are correlated traits, meaning that they change relative to each other. For example, down length and down production are closely correlated meaning one can measure down length and get an accurate estimate of production. This is good because it is much cheaper and easier to measure down length than it is to measure production.
Conversely, cashmere traits also show negative correlations or antagonisms between desirable traits. For example, cashmere weight and diameter are strongly negatively correlated meaning that selection for down weight alone will result in an increase in down diameter. Down production and live weight also exhibit a similar negative correlation.
Selection index design entails the calculations of EBVs for traits that have been identified as lacking and the combination of these traits into an index which has been weighted according to the defined objectives of the breeding program. Indices cannot be designed before the traits to be manipulated have been identified. According to Pattie and Restall (1996), Selection Indexes are used for two main purposes:
1) To combine information from all traits, taking correlations into account, to estimate an individual’s breeding value, and
2) To combine EBVs to provide an overall EBV or index score to rank animals. Economic weights may be used.
RESPONSE TO SELECTION: The amount of measurable response to a Breeding Program is influenced by four factors:
1) The amount of genetic variation within the population
2) The generational interval (how long it takes to completely turn over your genetic pool, ie: replace all animals within the population)
3) The selection intensity (the degree to which the females are identified as elite and used only with elite bucks. Progress will be faster within the elite doe herd than without).
The effective population size – Population size is important because it has an effect on the selection differential in that extremely superior animals are more likely to be found in a large population. Also, the variety of genetic makeup within a large population minimizes the chance of genetic inbreeding.