Behavior traits associated with career outcome in a prison puppy-raising program

Abstract

The Canine Behavioral Assessment and Research Questionnaire (C-BARQ©) (http://www.cbarq.org) has been used to measure behaviors associated with release or graduation from several assistance dog programs, however it has never been evaluated within a prison environment. This study investigated whether a modified version of the C-BARQ© can be utilized in a prison puppy-raising program (Puppies Behind Bars, PBB) to identify behaviors that are associated with dogs’ career outcomes. PBB dogs that successfully complete the program are placed as service dogs or explosives detection dogs (EDD). Dogs are released from the program as a result of behavioral or medical problems. The PBB program has more than one career outcome, facilitating an assessment of the C-BARQ© as a tool to identify specific working dog roles based on differences in C-BARQ© subscale scores. We examined the associations between subscale scores and career outcomes by comparing the scores of dogs with successful outcomes (service or EDD) with those released for behavioral reasons. We assessed the questionnaire’s application to the PBB setting and its ability to distinguish between outcomes. 314 paper copies of the C-BARQ© were completed by puppy raisers from seven correctional facilities in the New York area when their assigned dog was between 6 and 16 months old. Dogs that had successfully completed the PBB program or had been released due to behavioral issues were included, whereas dogs still in training and those released for medical reasons were excluded. A total of 271 completed C-BARQ© questionnaires were analyzed. Service dogs and EDDs were compared with released dogs to determine whether C-BARQ subscale scores were associated with outcome. Multinomial log-linear models containing one subscale score and fixed factors (age group, medical category, sex, neuter status within sex, the interaction between age group and medical category) and outcome as the dependent variable, were fitted for each subscale. Service dogs had lower stranger-directed aggression, owner-directed aggression, dog-directed aggression, dog-directed fear, dog rivalry, chasing, stranger-directed fear, and separation-related problems than released dogs. EDDs had lower trainability, dog-directed fear, dog rivalry, and attachment/attention-seeking behavior than released dogs. These findings suggest that some of the C-BARQ© subscales might be used in the future to predict outcomes for young dogs. Results show that the C-BARQ© can be applied to the PBB program; however, the omission of seven questions is recommended.