Can Citizen Science help assess donkey welfare using Qualitative Behaviour Assessment? How do visitors perceive donkey emotions?

Status
Status: 
Completed
Collaborators
Researchers
Details
When conducted: 
1 September 2016 - 31 July 2017
Country: 
United Kingdom
Core methodology: 
Stage 1: Video recordings made at The Donkey Sanctuary, Sidmouth. 19 videos selected to show a broad range of expressive repertoires across high and low energy as well as positive and negative valence. Stage 2: A Focus Group (FG) consisting of staff members identified 18 descriptors to be used. Stage 3: Two assessor groups (staff and visitor) observed and scored the expressive behavioural reactions of a focal donkey in each clip, using a Visual Analogue Scale (VAS). The data were then analysed in Minitab version 17 (State College, PA: Minitab, Inc.) using Principal Component Analysis (PCA) with a correlation matrix, assessing four components.
Aims: 
This study aimed to examine how visitors to an animal sanctuary viewed expressive behavioural states displayed by donkeys.
Results: 
According to the results from this study, both the staff and visitor assessor groups, when analysed separately as well as together, were able to use the QBA descriptors developed by the Focus Group (FG) to distinguish between different expressions observed in donkeys.
Conclusions: 

This is the first study on the use of Qualitative Behavioural Assessment (QBA) with donkeys being conducted through Citizen Science (CS). The results suggest that inexperienced visitors with little training can accurately score emotional states in donkeys in alignment with experienced staff members.

There was significantly strong agreement between the two groups of assessors (visitors and staff) when analysed separately. When both groups were pooled together, to create one all-assessor group, the agreement stayed significantly strong, indicating that both groups agreed with each other.

QBA is a useful tool in assessing positive welfare states in donkeys (of which there is little published data) and so is rightly placed in the AWIN welfare protocols alongside other welfare measurements for this species. This study supports other research which concludes that future development of QBA for a variety of species could include on-line training material, such as reference videos scored by experts and video teleconference instructions. It is through this that CS could have a role in QBA for organisations with webcams to record long-term measurements for observation of trends and monitor change. However it must be stressed that CS must not be used as a ‘cheap’ method of data collection, instead where appropriate it could be used to complement other measurements and must also be under routine quality assurance.

This initial investigation into QBA and CS with donkeys could be further investigated with other captive species, both wild and domestic. To be successful and scientifically accurate, QBA research into new species will need individual validation with species-specific descriptor lists.