In recent years, attention among dairy cattle geneticists and practica
l breeders has been drawn to the use of test (sampling) day records fo
r the estimation of breeding values and as an alternative formulation
of selection criteria as compared to records covering complete lactati
ons. The use of test day records offers the conceptual advantage of di
rectly considering records at their origin, i.e. at the day of recordi
ng. Under this approach genetic and environmental effects specific to
each single record can be considered. Results from the literature on t
he nature of environmental effects associated with test day records an
d on the estimation of genetic parameters for test day records are sum
marized. For the evaluation of animals, two different approaches exist
. The first concept consists of corrections for environmental effects
via statistical methods at the test day level. In a further step then
corrected records are aggregated into complete lactations for later pr
ocessing. Under the second approach statistical models are fitted incl
uding test day records directly. Records within lactation are taken as
repeated measurements. Results on the estimation of genetic parameter
s and breeding values under both approaches are discussed.