Validity and feasibility of qualitative behavior assessment for the evaluation of Nellore cattle temperament

Abstract

The aims of this study were to assess the validity and the feasibility of the qualitative behavior assessment (QBA) method as indicator of Nellore cattle temperament under field conditions, evaluating its associations with four other traditional methods and weight gain. The temperament and live weight of 2229 Nellore cattle was assessed at approximately 550 days of age. Five measurements of cattle temperament were recorded: flight speed test (FS, in m/s), visual scores of movement in the crush (MOV), crush score (CS), temperament score (TS), and the qualitative behavior assessment method (QBA), by using a list of 12 behavioral based adjectives as descriptors of temperament. Average daily weight gain (ADG) was calculated for each animal. For statistical analysis of QBA data, the Principal Component Analysis was used. A temperament index (TI) was defined for each animal using the scores for the first principal component. Pearson's correlation coefficients were estimated between TI with FS and ADG. A mixed model ANOVA was used to analyze the TI variation as a function of TS, CS, and MOV. The score plot for the first and second principal components was used to classify the cattle in four groups (from very bad to very good temperament). The first principal component explained 49.50% of the variation in the data set, with higher positive loadings for the adjectives 'agitated' and 'active', and higher negative loadings for 'calm' and 'relaxed'. TI was significantly correlated with FS (r=0.49; P<0.01) and ADG (r=-0.10; P<0.01). The means of ADG, FS, and the temperament scores (CS, TS, MOV) differed significantly (P<0.01) among the four groups, from very bad to very good temperament. The QBA method could discriminate different behavioral profiles of Nellore cattle and were in agreement with other traditional methods used as indicators of cattle temperament. Additional studies are needed to assess the inter- and intra-observers reliability and to study its association with physiological parameters. © 2013 Elsevier B.V.