This dissertation examines dog training in the United States to see how it is implemented and practiced and the potential ramifications for our pet dogs. In Chapter 2, word-based analysis compared the words practicing dog trainers use to describe their work and examined how gender and experience differ between methodologies. Given both the vast history of dog training and the variety of training information available to dog owners, in Chapter 3, I surveyed undergraduate students to determine where dog guardians obtain their information and how they might tackle dog problem behavior. Of 100 practicing dog trainers, only 20%, mostly female trainers using non-aversive methods, were certified. When asked who they asked for dog training advice, most dog guardians reached out to friends, family, or online sources and only about 20% ever attended a formal dog training class. In Chapter 4, dogs were walked on four different types of leash walking equipment which were attached to a strain gauge to assess if these equipment were any better or worse at reducing a dog’s pulling force; dogs pulled most on a martingale collar and there were no statistically significant differences among the other equipment types to each other. Through behavior analysis, none of the dogs in our study showed impaired welfare. In Chapter 5, I compared e-collar training to non-aversive methods in the ability to stop chasing of a lure. Dogs experiencing e-collar stimulation were successful in stopping chase of the lure both during training sessions and in testing. None of the dogs in either non-aversive group was successful in the tests. The findings did not indicate that dogs in the e-collar group, or the non-aversive groups, differed in welfare. As in Chapter 4, despite the lack of immediate negative welfare effects for the dogs experiencing aversive methods like the e-collar, given the demonstrated evidence for Chapters 2 and 3 that individuals who use these aversive tools, either practicing trainers or the general public, may lack the foundation to properly implement them, I recommend caution in their public sale and do not advocate for their use.

Each year, nearly three million dogs will enter one of over 13,000 animal shelters in the United States. The purpose of this dissertation is to better understand how breed identity and dog welfare in the shelter, in addition to post-adoption and fostering interventions out of the shelter, can contribute to the betterment of dog lives. In Chapter 2, I conducted the largest sampling of shelter dogs’ breed identities to-date to determine their breed heritage and compare shelter breed assignment by staff as determined by visual appearance to that of genetic testing. In Chapter 4, I examined the efficacy of a post-adoption intervention intended to reduce returns by encouraging physical activity between adopters and their dogs. In Chapter 6, I examined the effects of brief stays in a foster home on the urinary cortisol: creatinine ratios of dogs awaiting adoption compared to ratios collected before or after their stays; and in Chapter 7, I characterized the relationships between multiple physiological, health, and cognitive measures and the in-kennel behavior of shelter dogs.

Four suggestions from the findings of this dissertation that will likely better the lives of dogs living in animal shelters are: 1) Shelter dog breed heritage is complex and visually identifying multiple breeds in a mixed breed dog is difficult at best. Shelters should instead focus on communicating the morphology and behavior of the dogs in their care to best support adopters. 2) While encouraging walking did not influence owner behavior, adopters who reported higher obligation and self-efficacy in dog walking were more active with their dogs. Thus, post-adoption interventions that can effectively target owner perceptions of obligation and self-efficacy may be more successful in changing behavior. 3) Temporary fostering is an impactful intervention that reduces stress for dogs awaiting adoption; however addressing stressors present at shelters that are likely contributing to higher stress responding is also needed. 4) It is possible to predict the internal stress responding of shelter dogs by observing their overt, in-kennel behavior, and this study is a first step in assessing and improving the welfare of dogs living in animal shelters.