Author
 |
POWELL, REX |
|
NORMAN, H |
|
Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 6/21/1999 Publication Date: N/A Citation: N/A Interpretive Summary: Semiannual genetic evaluations for dairy cattle yield in the US were replaced by quarterly evaluations beginning in May 1997 with the stipulation that there would be a review of the decision in 1998. The more frequent results caused added marketing effort for semen suppliers and was not universally welcomed. This study was initiated at the request of the Council on Dairy Cattle Breeding to examine the genetic consequences of quarterly evaluations. Holstein bulls reported through the National Association of Animal Breeders were studied for amount of change in evaluations at various intervals, ability to identify bulls changing substantially in estimated merit, and usefulness of identifying superior new bulls sooner. Quarterly evaluations changed about 30% less than semiannual evaluations which was essentially as expected based on theory. Bulls with large changes over a 6-month interval were more accurately estimated at the 3-month evaluation 94-96% of the time. Evaluations for superior bulls receiving evaluations 3 months sooner due to the quarterly schedule were, as expected, not as accurate compared to the second evaluations having additional data. However, they were considerably more accurate than parent average that would otherwise have been the best genetic information available. Timeliness in identifying these top bulls has many advantages to the dairy industry. Quarterly evaluations are a valuable resource and their continuation was not only supported by the Council, in part as a result of this study, but the International Bull Evaluation Service has subsequently decided to likewise move from two to four evaluations per year. Technical Abstract: Genetic evaluations of Holstein bulls for February 1997 through May 1998 were examined for the value of more frequent evaluations in more timely identification of bulls with changing evaluations and new superior bulls. Changes between evaluations with half the interval were reduced by 30 percent. About two-thirds of the evaluations at a given run were closer to the evaluations 3 mo later than were evaluations 3 mo earlier. However, improvements in accuracy were 94-96 percent for the subset of bulls changing substantially from 3 mo before the subject evaluation to 3 mo after. With quarterly evaluations, half of bulls have first PTA available 3 mo sooner than with semiannual evaluations. These evaluations were clear improvements over the parent averages that would otherwise have remained the best genetic estimates for 3 mo more. Correlations of parent average with PTA about a year later were .5 to .6 whereas the initial PTA were correlated about .8 with the later PTA compared to .9 for the second PTA. Although later data are generally expected to be improved estimates of true merit, the timely results provided by quarterly evaluations were useful in identifying bulls with PTA changing substantially and in identifying top new bulls. |
