Text Mining Analysis to Evaluate Stakeholders’ Perception Regarding Welfare of Equines, Small Ruminants, and Turkeys

SIMPLE SUMMARY: Consumers are currently more sensitive regarding the way animals are kept and handled and, in general, have increased their awareness towards animal welfare. There is a close link between welfare of animals and stakeholders’ perception of their needs. This study aimed at investigating stakeholders’ perception of the welfare of equines, small ruminants, and turkeys using a text mining approach to analyze their answers to open-ended questions. A total of 270 surveys were collected from respondents of 32 different countries. Independently from the species, respondents considered that an animal needs appropriate nutrition to be fit, healthy, and productive. The “environment” was considered particularly important for turkey stakeholders, whereas housing factors were relevant for goat stakeholders. Horse stakeholders also considered “exercise”, “pasture” and “proper training” important. Although the sample was too small to analyze variation in welfare perception among stakeholders of different species, text mining analysis seems to be a promising method to investigate stakeholders’ perception of animal welfare, as it emphasizes their real perception, without the constraints deriving by close-ended questions. ABSTRACT: Welfare of animals significantly depends on how stakeholders perceive their needs and behave in a way to favor production systems that promote better welfare outcomes. This study aimed at investigating stakeholders’ perception of the welfare of equines, small ruminants, and turkeys using text mining analysis. A survey composed by open-ended questions referring to different aspects of animal welfare was carried out. Text mining analysis was performed. A total of 270 surveys were filled out (horses = 122, sheep = 81, goats = 36, turkeys = 18, donkeys = 13). The respondents (41% veterinarians) came from 32 different countries. To describe welfare requirements, the words “feeding” and “water” were the most frequently used in all the species, meaning that respondents considered the welfare principle “good feeding” as the most relevant. The word “environment” was considered particularly important for turkeys, as well as the word “dry”, never mentioned for other species. Horses stakeholders also considered “exercise” and “proper training” important. Goat stakeholders’ concerns are often expressed by the word “space”, probably because goats are often intensively managed in industrialized countries. Although the sample was too small to be representative, text mining analysis seems to be a promising method to investigate stakeholders’ perception of animal welfare, as it emphasizes their real perception, without the constraints deriving by close-ended questions.