Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

The qualitative behavioural assessment of sows under different group housing conditions

Clarke, Taya (2015) The qualitative behavioural assessment of sows under different group housing conditions. PhD thesis, Murdoch University.

PDF - Whole Thesis
Download (2MB)


In response to animal welfare concerns surrounding the confinement of sows, the Australian pig industry made the voluntary commitment to group house all sows as of 2017. As the industry changes, so too does the requirement for welfare assessment tools. This study tested the application of Qualitative Behavioural Assessment (QBA) as a method of assessing sow welfare under commercial piggery conditions. Video footage was collected of sows at the point of mixing into group pens of different design. The behavioural expression of these sows was compared using QBA to investigate changes in qualitative behavioural expression over time, assess the effect of housing conditions and parity, the relationship with reproductive success, as well as comparing application of two QBA methodologies. Footage was also analysed quantitatively to identify changes in activity budgets over time for comparison with QBA scores. In the first 10 minutes after mixing into groups with free-access feeding stalls, sows were more ‘active/agitated’ and there was more investigative behaviour compared with 60 minutes after mixing, when sows were more ‘bored/calm’ and spent more time lying down. When mixed into pens with a concrete partition, sows were described as more ‘calm/relaxed’, walked less, and laid down sooner, compared to sows in open pens (who were more ‘aggressive/tense’). There was no relationship found between the QBA scores of individual sows at mixing and their reproductive success during that parity. When moved into mixed-parity groups, parity 2 sows were described as more ‘curious/interested’, parity 4 sows were more ‘anxious/upset’ and parity 6 sows were more ‘calm/subdued’. During the first feeding cycle in pens with an electronic sow feeder (ESF), sows were more ‘interested/alert’ in the early and the late stages of the 24-hour cycle and the demeanour of sows was more ‘relaxed/calm’ and there was less activity around the feeder after 7 days post-mixing compared to sows at their first feed as a group. When the same sows were re-penned 6 weeks post mating, they were more ‘bored/submissive’, investigated the feeder more and lay around the feeder less compared with the same sows at mixing. Comparing Fixed Lists (FL) and Free Choice Profiling (FCP), both methods showed significant inter-observer reliability, demonstrating that, regardless of whether they are given a fixed list of descriptive terms (FL) or are allowed to generate their own (FCP), observers score sow behaviour in a similar way. This study found that QBA is an effective tool in not only detecting changes in behavioural expression over time, but also demonstrated that it is applicable in an industry setting for comparing housing options. Both the FL and FCP approach to assessing the qualitative behavioural expression of sows could be applied under commercial piggery conditions.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): School of Veterinary and Life Sciences
Supervisor(s): Fleming, Trish, Miller, David, Collins, Teresa and Pluske, John
Item Control Page Item Control Page


Downloads per month over past year