Biases in Modified Contemporary Comparisons due to Year of Calving

Open Access
Publisher Website
Google Scholar
Abstract
Yearly sire summaries were calculated on each of 63 Holstein bulls with more than 3000 daughters. Each summary included all records on those daughters with a particular year of 1st freshening. A minimum of 100 daughters/yr was required before summaries were included in the analysis. Modified contemporary deviations and approximated herdmate deviations were compared. Multiple regression analysis showed that year of 1st freshening accounted for .1% of the variation in modified contemporary deviation but accounted for 4.3% of the variation in herdmate deviation when year of freshening was the only variable in the model. Sires accounted for a greater percentage of the total variance among modified contemporary deviations (95.4%) than among herdmate deviations (89.7%) when included as the only variable in the model. When sire and year of 1st freshening were examined, the regression of modified contemporary deviation on year was -6.4 kg while that for herdmate deviation was -26.7 kg. Corrections in modified contemporary deviations for genetic merit of contemporary sires appeared to have removed a substantial portion of the bias that was in the herdmate comparison as the result of genetic trend.