Same Pig, Different Conclusions: Stakeholders Differ in Qualitative Behaviour Assessment

Abstract

Animal welfare in pig production is frequently a topic of debate and is sensitive in nature. This debate is partly due to differences in values, forms, convictions, interests and knowledge among the stakeholders that constitute differences among their frames of reference with respect to pigs and their welfare. Differences in frames of reference by stakeholder groups are studied widely, but not specifically with respect to animal behaviour or welfare. We explored this phenomenon using a qualitative behaviour assessment (QBA). Participating stakeholders were classified into two expert groups consisting of pig farmers (N = 11) and animal scientists (N = 18) and a lay-group consisting of urban citizens (N = 15). The stakeholders were asked to observe the behaviour of a specific pig in each of the nine videos and to assign a score for each video using 21 predefined terms describing the mood, such as 'happy' or 'irritated'. They were asked to complete two additional questionnaires to obtain information on their frames of reference. Results from the QBA showed -014-9513-z that the pig farmers observed the behaviour of pigs more positively than the urban citizens and the animal scientists. This was evident from the consistently higher scores on the positive terms to assess pig behaviour. The questionnaires revealed that the farmers had a different frames of reference regarding pigs and different understanding of welfare, which might explain the differences in assessment. In a follow-up stakeholder workshop, which focussed on differences in observation, QBA showed to be an effective tool to stimulate mutual learning among stakeholders, which is necessary to find shared solutions.123 J Agric Environ Ethics (2014) 27:1019-1047 DOI 10.1007/s10806

Are observer ratings influenced by prescription? A comparison of Free Choice Profiling and Fixed List methods of Qualitative Behavioural Assessment

“…We found similar significant interobserver reliability for FL and FCP groups and significant agreement in GPA dimension 1 and 2 scores between the two observer groups, suggesting that observers in each group scored the body language of the sows in a similar way. Clearly, therefore, selection of either FL or FCP methodology cannot explain discrepancies in conclusions about reliability of observer scores between previous QBA studies Duijvesteijn et al, 2014). Greater understanding of the methodological and statistical procedures will help elucidate the benefits and potential issues associated with using either FL or FCP methods for qualitative behavioural assessment.…”

“…Wemelsfelder et al (2012) demonstrated consistency and agreement in how pig farmers, veterinarians and animal activists scored pig behaviour, but by contrast, Duijvesteijn et al (2014) reported that pig farmers observed the behaviour of pigs in a more positive way compared to animal scientists and urban citizens. Duijvesteijn et al (2014) used a FL methodology (the same descriptive terms as the present study) while Wemelsfelder et al (2012) used FCP. It is argued that these methodological differences accounted for different outcomes of these two studies (Duijvesteijn et al, 2014).…”

“…Recent reviews suggest that observer-based methods can be robust and perform a useful task in scientific investigations (Meagher, 2009;Whitham and Wielebnowski, 2009), although other authors have questioned the validity of rater assessments due to a lack of inter-observer agreement (Bokkers et al, 2012). Wemelsfelder et al (2012) demonstrated consistency and agreement in how pig farmers, veterinarians and animal activists scored pig behaviour, but by contrast, Duijvesteijn et al (2014) reported that pig farmers observed the behaviour of pigs in a more positive way compared to animal scientists and urban citizens. Duijvesteijn et al (2014) used a FL methodology (the same descriptive terms as the present study) while Wemelsfelder et al (2012) used FCP.…”

“…Duijvesteijn et al (2014) used a FL methodology (the same descriptive terms as the present study) while Wemelsfelder et al (2012) used FCP. It is argued that these methodological differences accounted for different outcomes of these two studies (Duijvesteijn et al, 2014).…”

The contribution of qualitative behavioural assessment to appraisal of livestock welfare

“…For example, Duijvesteijn et al (2014) found that, although observers used terms similarly (largely the same set of descriptive terms were correlated with each of the behavioural dimensions) and had the same intra-observer correlations (each observer saw the same video clip twice), farmers tended to score the behavioural expression of animals more positively than urban citizens or animal scientists. Bokkers et al (2012) compared the scores attributed to the same footage by eight experienced observers for each descriptive term over two viewing sessions, separated by 9 months.…”