What are EBVs?


EBV is the acronym for "Estimate Breeding Value." An EBV is a numerical estimate for the genetic value of an animal. EBVs are scientifically-based and industry-proven. They are the best way to compare the genetics of livestock in different flocks/herds. They are the most accurate way.

For those familiar with EPDs in beef cattle, EBVs are very similar. Where EPDs predict the genetic merit of an animal's progeny, EBVs predict the genetic merit of the animal itself. An EBV is equal to twice the EPD. We use EBVs to predict the performance of an animal's offspring.

Most economically-important traits cannot be determined by simply looking at an animal. It is particularly hard to determine which animals have superior maternal traits. Traits need to be measured. EBVs can be calculated for any trait that can be measured. There are EBVs for reproductive, growth, carcass, wool, and fitness traits.

Reproductive traits include number of lambs/kids born (NLB), number lambs/kids weaned (NLW), ewe productivity index (EPT), and scrotal circumference (SC). Growth traits include birth weight (BWT), maternal weaning weight (MWWT), weaning weight (WWT), post-weaning weight (PWWT), and yearling weight (YWT). Carcass traits include post weaning eye muscle depth (PEMD) and post weaning fat depth (PFAT). Wool traits include fleece weight (GFW), fiber diameter (FD), and staple weight (SL). Fitness traits include fecal egg count (WFEC and PFEC). They're always working on more traits.

EBVs are expressed in the same unit as the trait is measured. A weaning weight EBV of +3 means that an animal's weaning weight is 3 kg heavier than the breed average. The animal will pass a +1.5 kg advantage to his offspring since offspring receive 50% of their genes from each parent. The units are kilograms (kg) since the data is processed outside of the US.

Sometimes a positive number is more desirable (e.g., weaning weight), sometimes a negative number is more desirable (e.g., fecal egg count). A fecal egg count EBV of -50% means the animal's fecal egg count is 50% less than the breed average. Some EBVs are expressed as percentages (e.g., number lambs/kids born). A number of lambs born EBV of 110% means the animal will produce 10% more lambs thant the breed average.

A very important aspect of EBVs is accuracy. Accuracy is a measure of confidence in an animal's EBV. Each EBV has an accuracy value. Accuracy values are on a scale from 0 to 100. 100 represents the animal's true breeding value. Many factors influence accuracy, including the amount of information for the animal, the amount of information for the relatives, the heritabilty of the trait, and number of animals being compared (contemporaries). The higher the accuracy, the more confident we are in the EBV.

Submitting data
EBVs start with collection of the data on the farm/ranch. There are certain guidelines that must be followed for data collection. In the US, the raw data is submitted to the National Sheep Improvement Program (NSIP; nsip.org). It is entered into software called Pedigree Master (free). You can upload data from a spreadsheet into Pedigree Master. The EBVs are then calculated by Sheep Genetics of Australia.

Once the data is received, it is adjusted for known non-genetic influences such as age of dam, type of birth and rearing, and age of lamb. It is very important to remove the environmental influences on performance. The performance of the animal is then compared to the average performance of the other animals in its contemporary group.

A contemporary group is a group of animals of similar age that have been managed the same. Orphans would not be in the sample contemporary group as dam-raised lambs/kids. If you separate ewe/does raising triplets for preferential feeding, their offspring would be in a different contemporary group than the twins and singles. At least 10 to 20 animals are needed (per sire) in a contemporary group.

The performance record for each animal is compared to the performance records of its relatives in proportion to how closely the animals are related. The numbers are then adjusted for heritability and number of records available. While this sounds complicated, all the math is done by a computer.

Some people wonder how you can compare animals that have been raised differently. Won't performance be higher in sheep/goats that are fed more? Yes, it probably will be, but EBVs look at the differences in performance. You are able to compare sheep/goats raised in different production environments by evaluating the same genetics in different production environments. Farm A uses a ram raised by Farm B. Farm B uses a ram raised by Farm A. Farm C uses a ram from both farms (A,B). If there aren't genetic linkages between the flocks/herds, EBVs are only able to compare performance within the same flock/herd. For EBVs to be most useful (across-flock comparisons), producers need to make efforts to share genetics.

Using EBVs
When using EBVs to select breeding stock, it's important to use them properly. It's best to select animals with "balanced" EBVs. This means selecting an animal that has favorable EBVs for most of the traits important to you and close to average EBVs for the traits less important. Another way to select for balanced EBVs is to use one of the various selection indexes.. Selection indexes combined EBVs for different traits into a single EBV. There are indexes for Katahdins, Polypays, terminal sire breeds, and western range breeds. Single trait selection is not recommended, when using EBVs or any other selection strategy. Other animal industries have made mistakes doing this.

If you are a commercial producer, even if you do not submit data to NSIP, it is a good idea to purchase breeding stock, especially rams, with EBVs for the traits that are imporant to your flock. You can purchase rams from sales which provide EBVs or you can purchase directly from flocks that are enrolled in NSIP. Breeds with the most participation in NSIP include Katahdin, Polypay, Suffolk, and Targhee. Any breed can submit data to NSIP. The sheep don't need to be registered, but they should be purebred. NSIP doesn't have a way to compare crossbred animals (yet). Each breed "customizes" its EBVs. There is an flock enrollment fee and per head charges for EBVs.

Unfortunately, there are very few goat producers that are enrolled in NSIP. Meat goat producers tend to prefer Central Performance "Buck" tests. Dairy goat producers have DHIA. Using the data from central performance tests is a much less accurate way to identify genetically superior animals. Only the data from one animal is being considered, whereas EBVs include data from relatives, other farms, and other correlated traits. Plus, performance tests don't evaluate maternal traits.

EBVs are the best way to compare animals from different flocks/herds. Animals with EBVs aren't necessarily better than those without EBVs. Their genetics are simply quantified; thus, more predictable. When you buy a sheep/goat without EBVs, you are buying a unknown commodity. It could be a good one or it could be a bad one. Think about that before writing a big check for ram or buck.


Additional reading

National Sheep Improvement Program
Understanding Sheep Estimated Breeding Values - Katahdins.org
Frequently asked questions about estimated breeding values and NSIP
Glossary of terms for NSIP members and users of EBVs
Ram Buying Guide: Using EBVs to select quality rams & reach production goals
The Breeder's Guide to LambPlan and KidPlan