Analyzing the Safewalk Program with SAS: Saving Shelter Dogs One Walk at a Time

This 2013 study from the Massachusetts Society for the Prevention of Cruelty to Animals (MSPCA) details their recent success with “Safewalk,” a program designed to train staff on how to behave around shelter dogs to help the dogs survive shelter life and increase their chances of adoption. The paper explores not only why and how Safewalk was implemented, but how initial data from the program seemed to indicate that it was actually increasing euthanasia rates and length of stay at the shelter, which later turned out to be false. The paper pays special attention to the case study of Chopper, a dog who was surrendered twice to the shelter due to behavioral problems, but was eventually homed successfully after Safewalk was implemented.

In a bid to lower euthanasia rates and increase the speed and rate of adoptions, the MSPCA implemented a program called Safewalk. Rather than training dogs, however, the program was meant to train the humans at the shelter, specifically staff and volunteers, on how to behave around dogs residing there. The program was put in place to replace a previous system that caused friction between staff and volunteers, and did not seem to improve the shelter dogs’ lives in the process. In short, Safewalk would combine human training with a system of color-coding the dogs based on behavioral traits, to help the dogs adjust to shelter life, and get adopted more readily.

Though the staff and volunteers intuitively felt that Safewalk was working, especially for more traditionally hard to adopt breeds like pit bulls, the data was showing something else: length of stay for the pit bulls was increasing, and euthanasia rates were rising. As the numbers seemed to contradict what staff and volunteers were experiencing on the ground, researchers dug deeper into the data and found that the implementation of the program coincided with when the MSPCA became the only “open” shelter in Boston. In other words, the MSPCA was the only place where other shelters would send dogs that were otherwise considered “unadoptable.” In this extra analysis into the context of the data, researchers also found that Chopper, the dog that inspired the Safewalk program in the first place, was an outlier in the statistics and skewed the averages for length of stay greatly. Once these factors were accounted for, the data began to make more sense, and became much more encouraging.

Researchers outline the two biggest takeaways from the study. First and foremost, “with consistent training for humans, pit bulls [and other dogs] can withstand being homeless for longer periods, and get adopted at higher rates.” Safewalk, as a training program, was a contributing factor in seeing some of the longest residents at the MSPCA, including Chopper, adopted out successfully and not returned a few months later. Secondly, and of equal importance in this case, what the authors refer to as the “Know Thy Data” rule. Without knowing the context of the MSPCA’s transition to an open shelter around the same time as Safewalk was implemented, it may have appeared that it was a failed program that was actually hurting shelter dogs. Instead, researchers were able to appropriately analyze the data and get a proper sense of Safewalk’s success.

Original Abstract The MSPCA (Massachusetts Society for Prevention of Cruelty to Animals) in Boston initiated the Safewalk® Program in January 2009. This program was designed to enrich the experience of shelter dogs by providing training to volunteers and staff that allow dogs of varied backgrounds and temperaments to be exercised safely, as well as promoting behaviors encouraging adoption on the adoption floor. Intuitively the program was felt to increase adoption rates, particularly in “hard to adopt” dogs, but proving the benefits of the program was a challenge, especially given the greatly increased numbers of homeless dogs during the Safewalk Program time period. The MSPCA has recorded statistics on all animals in their care compiled with Chameleon© software. Records for all dogs for 30 months prior to the initiation of Safewalk and 30 months after were selected from the Chameleon database. This data extract was analyzed using multiple SAS procedures in SAS/STAT. This paper will demonstrate how SAS analysis, output and statistical graphs allowed us to assess the effects of the Safewalk Program and which populations it most affected.