221 Late-Breaking: The relationship between self-reported and expert scored measures of skill in cattle handling.

Abstract Few matters of great societal concern transcend that of a common requirement for food to survive; while everyone eats, a mere 1.36% of the U.S. workforce is employed in agriculture, which includes crop and animal production. The importance of occupational skill has long been recognized, yet attributes of cattle handling skill remain largely unresolved in the literature. For animal caretakers, there is no industry standard defining skill level, and few barriers to entry, such as requisite training. Instead, this sector of the workforce has traditionally relied on self-reported years of experience (YOE), often gained growing up on a farm or ranch, as a proxy for skill. When 50% of the population was involved in agriculture, this may have been a realistic prerequisite, but for modern agriculture, the assumption that familiarity with agriculture is a prerequisite can be deadly. While fatal injuries in crop production decline, fatal injuries steadily increase for animal workers, despite adoption of technology by both sectors. The relationship between skill level of cattle handlers and risk for injury has not been well described in the literature. We compared self-reported YOE, self-rated expertise, and self-reported scores for ten cattle handling tasks with independently assigned scores, and investigated the relationship between demographic information, prior injury history, self-reported measures, and expert assessment of skill level. A panel of experts identified ten tasks intended to be representative of a core skill set used by workers across sectors of the cattle industry. Each task included categories reflecting attributes of skill believed to be important for both animal welfare and handler safety. The skill assessment tool (SAT) developed by the panel was used to evaluate 15 participants of varied backgrounds. Participants were asked to report experience in number of years for all major livestock species; all reported ≥5 YOE. Participants were asked about prior injuries; seven reported no previous injuries; eight reported previous injuries sustained working with livestock. Participants completed 10 tasks with a group of cattle and were asked to score their own performance for each task. Classroom training covering principles of low-stress cattle handling was provided between handling trials. Following the training, participants repeated the cattle handling tasks and again scored themselves. The same expert was present to score all handling trials live. Video recordings were used post-hoc to validate live scores. Demographic information and self-reported scores were compared to expert-assigned scores, and the SAT was sensitive enough to detect differences in skill level between participants for both self-reported and expert-assessed scores. Male participants scored themselves higher than females, and all participants scored themselves higher than the expert. No differences were observed for self-reported or expert assigned scores before or after the training, suggesting this approach to training has limited efficacy.