Investigating the ability of standard tests to detect improvements in behavior after a 9-week training program in shelter dogs - A pilot study

Abstract

Annually, millions of dogs enter shelters worldwide. As such, it is vital to develop effective methods for assessing adoptability and promoting successful rehoming that can be easily implemented by organizations with minimal resources. Many shelters use temperament testing and training programs to facilitate these goals, but there is limited research on the effects of training programs on shelter dog behavior, and no standardized tests for assessing the skills imparted by those programs. Therefore, the primary goal of this study was to evaluate the sensitivity of temperament and basic education tests to detect changes in dog behavior after training in a shelter context. As a secondary goal, we aimed to compare video and live scoring of dog behavior in these tests. Twelve shelter dogs were enrolled in a 9-week training program to learn basic education skills (e.g., walking on a leash without pulling, not jumping on people, responding to cues such as ‘sit’ or ‘stay’). Both before and after the program, each dog was tested using a previously evaluated and published temperament test designed for use in shelters and a test of basic education skills developed by the research team to determine the ability of the tests to measure training progress. Results showed no differences between pre- and post-training temperament test scores for the categories Human Sociability and Playfulness nor for the total temperament test score; however, a decrease was observed in the score for Dog Sociability. Regarding the basic education test, results showed improvements in scores for the categories ‘Walking on a loose leash’, ‘Sit, down and staying in place’, ‘Coming when called’ and for the total basic education test score; conversely, no differences were found for the categories ‘Accepting a friendly stranger’, ‘Waiting politely for petting’, ‘Walking through a crowd’ and ‘Supervised Separation’. Moreover, we found an overall high interrater agreement for both tests. The outcomes of the present study suggest that i) the proposed basic education test is a reliable measure for detecting improvements imparted by basic education programs, and that ii) basic education training has minimal effects on behavior assessed in the employed temperament test. We discuss how the utilized behavior tests and training programs may be beneficial in a shelter context.