Skip to main content
ARS Home » Research » Publications at this Location » Publication #78704
Title: COLLECTION AND MANAGEMENT OF TEST DAY DATA IN THE UNITED STATES

Author
item Wiggans, George
item Lake, Jill
item Van Tassell, Curtis - Curt
item Vanraden, Paul

Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: 6/1/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Data from about 100 million cow test days have been collected for calvings since Jan. 1, 1990, and stored in data tables based on cow, lactation, and herd; cow and test day; and herd and test day. Those data were collected from records provided for USDA-DHIA genetic evaluations and from archive files contributed by several dairy records processing centers and universities. Only test days for lactations that met current editing requirements were retained. Archive files were necessary to complete test day data prior to Nov. 1996. Test day data were nearly complete for all states except Michigan, for which monthly submission of records in progress began during Jan. 1994 and no archive data were available. Some archive data did not include an indication of times milked on test day; therefore, times milked on test day was inferred from records in progress of other cows for that test day. Because cows sometimes changed herds and because most of the archive data were stored based on lactation or cow, an incorrect herd code sometimes was associated with a test day record. The data table based on herd and test day included the number of cows with yield on that test day, which allowed herd code errors to be detected. A new format was developed for routine transfer of all test day data for a lactation. Procedures for routine extraction of new data and updating of data tables are being developed. Test day data collected since 1990 are adequate for estimation of variance components and development and implementation of a genetic evaluation system based on test day yield. However, lactation data before 1990 should also be included to account for previous selection and genetic trend.