Tip of The Day - October 18, 2016

Photo by Tom Semadeni

March 3

1988 Repeated Measures in Station Tests

Paola de Rose and Jim Wilton

Six measurements, one every 28 days, of weight (WT), backfat (BF), scrotal circumference (SC) and height (HT) were taken on samples of beef bulls on station tests, starting at 200 days of age. Growth patterns were examined. By start of the test, bulls of all breeds had already achieved 55%, 70%, and 90% of their end of test levels for WT, SC, and HT. The majority of growth occurs prior to weaning.

Measurements taken in the bull test program were highly repeatable, with the exception of BF which may require at least two measurements for an acceptable level of repeatability. All growth patterns are essentially linear with respect to age, within the age range studied (200-380 d). There was evidence of differential age effects across test groups and breeds. Because growth of SC and HT was slight, there is little point in including them in post-weaning analyses. WT and BF can be based on only start and end of test measures with a high degree of reliability. The use of individual animal regressions to obtain yearling values is not justifiable based on correlations with the best method.

Note: This study begs for a random regression analysis, but random regression models were not discovered for another six years. However, the conclusions were valid, in that, measures at start and end of test were all that is necessary. No need to use labour to take weights every 28 days, and no need to disrupt the animals at the same time. And no need for random regression analyses, in this situation.

Photo by Tom Semadeni

Moose