A note on: On-farm testing of the behavioural response of group-housed calves towards humans; test-retest and inter-observer reliability and effect of familiarity of test person

In this study, a test of the behavioural response of group-housed calves towards humans has been evaluated regarding inter-observer and test-retest reliability as well as effect of familiarity of test person. On seven commercial farms, 186 group-housed calves at the age of 30-183 days were tested when fed concentrate in their manger. A test person approached the calves individually and tried to touch their shoulder and head. The behavioural response of the calves was categorised 0-3 as either 'withdrawal' or 'no withdrawal' from approach of test person, shoulder touch, and head touch, respectively. One hundred and thirty-four calves were tested on two consecutive days, and 166 calves were tested towards both the stock person and an unfamiliar test person. Two observers simultaneously recorded the test response of 144 calves when tested towards the stock person. The inter-observer agreement was high (weighted kappa coefficient 0.77) and the test-retest reliability relatively high (weighted kappa coefficient 0.62). The calves seemed to distinguish between the stock person and the unfamiliar test person in their test response; in one farm more calves avoided the familiar test person (P < 0.05), in another farm more calves avoided the unfamiliar test person (P < 0.05). (c) 2005 Elsevier B.V. All rights reserved.