Avoidance distance in sheep: Test–retest reliability and relationship with stockmen attitude

Validity and reliability of simple measures of human animal relationship, such as avoidance distance in the pen (AD), avoidance distance at the manger (ADM) and percentage of animals not accepting to be touched at the manger (ADM0), were assessed in 20 sheep farms. Validity was tested using an attitude questionnaire administered to the stock-people and calculating the correlation of questionnaire scores with AD, ADM and ADM0. Convergence among these three variables was also computed. Short-term test–retest reliability was evaluated correlating the measurements recorded in the two visits performed in each farm. Stock-people rated the four sections of the questionnaire at scores above the central point, thus indicating that the attitude of stock-people was generally positive. AD and ADM were higher than reported in other dairy ruminant species. AD was significantly higher than ADM (P < 0.001). A high correlation was observed between ADM and ADM0 (r = 0.810; P < 0.001), whereas lower correlations were observed between AD and ADM (r = 0.445; P < 0.05) and AD and ADM0 (r = 0.570; P < 0.01). ADM and ADM0 showed significant correlation with the questionnaire total score (r = −0.435 and −0.509, respectively; P < 0.05), whereas AD was unrelated with the questionnaire total score. A good test–retest reliability was observed for all the three measures (P < 0.001), although ADM0 showed a lower coefficient value (r = 0.679) in comparison with AD (r = 0.898) and ADM (r = 0.845). The marked gregarious behaviour exhibited by sheep in enclosed areas (when approached in the pen animals may run towards the observer) may impair AD feasibility in terms of observer safety. The present results indicate the measures gathered at the manger as the most suitable for the assessment of human–animal relationship in sheep.