Conference proceedings article
How to measure dairy cows´ responsiveness towards humans in breeding and welfare assessment?

Publication Details
Ebinghaus, A.; Ivemeyer, S.; Lauks, V.; Santos, L.; Brügemann, K.; König, S.; Knierim, U.
De Jong, Ingrid; Koene, Paul
Wageningen Academic Publishing
Publication year:
Pages range:
Book title:
7th International Conference on the Assessment of Animal Welfare at Farm and Group Level


In dairy cattle breeding and animal welfare
science uncertainties exist regarding the quality of measures of cows’
responsiveness towards humans. This study investigated criterion validity of four
behavioural measures applied in the barn – avoidance distance (AD), tolerance
to tactile interaction (TTI), release behaviour (RB), and qualitative behaviour
assessment (QBA) in a handling situation 1.) in comparison with the existing behavioural
breeding trait of milking temperament (MT) at animal level, and 2.) in
comparison with cows’ behaviour during milking at herd level. On 33 farms we collected
data on AD, TTI, RB, QBA, as well as MT that is routinely scored by the farmer,
and recorded the number of steps and kicks per cow during milking.

Inter-test associations on animal and herd
level were investigated by Spearman rank correlation analysis. While at animal
level AD and TTI correlated weakly (rs=0.28, p<0.01, n=1890), among
the other behavioural measures moderate to strong correlations were found (rs=0.46-0.74,
p<0.01, n=582-1890). Between MT and any other measure no significant
correlation was found. At herd level, medians of AD and QBA, means of steps and
kicks, and percentages of cows that were scored nervous regarding TTI, RB, and
MT were calculated. Significant correlations were found between AD, TTI, RB,
and QBA, in general higher than at animal level, and between steps and kicks (rs=0.80,
p<0.01, n=24). But neither steps and kicks nor MT were correlated with AD,
TTI, RB, or QBA.

Associations of different strengths between AD,
TTI, RB, and QBA suggest that the measures partly reflect different aspects of
cows’ responsiveness towards humans, possibly depending on former general as
well as individual experiences. The absence of correlations with the behaviour
during milking underline methodological problems with MT, but also question the
relation between behaviour in the milking parlour and responsiveness in the

Last updated on 2020-23-07 at 13:07