In his paper ‘Temperament Testing in the Age of No Kill’ for Best Friends Animal Society, Nathan Winograd discussed the limitations of temperament testing in the US, namely that it wasn’t very scientific;
Temperament testing is a series of exercises designed to evaluate whether an animal is aggressive. Because dog behavior is highly specific to context, it is unfortunately not enough to say that a dog is friendly and of reasonably good temperament if she comes into a shelter with her tail wagging. The flip side is also true. Because the shelter is a highly stressful, unnatural, and frightening environment for a dog who has just been abandoned by a family, the fact that a dog is scared and growls at staff on intake is not enough to make a determination that the dog is unfriendly and vicious. So it is not only fair, but a good idea, for shelters to evaluate dogs to make sure they can safely be placed into loving new homes.
But temperament testing has many limitations. It requires skill and training; the results greatly depend on the environment in which the test is conducted; and, because its predictive validity has not been established by any stretch, it can – and often does – result in dogs being wrongly executed.
However, despite their limitations, temperament tests are still a useful tool in a No Kill setting; as when using them as a guide, rather than a pass/fail for euthanasia, the process of determining which rehabilitation a dog needs can begin.
Dog behavior runs the gamut from simple bad manners, such as jumping up on people, to global undersocialization. Some behaviors are easily remedied; others are beyond the ability of a shelter to rehabilitate. Certainly no shelter should kill a dog for bad manners, and a no-kill shelter is obligated to rehabilitate all treatable behaviors, even those like food aggression.
But the situation in most pounds (and some shelters) across Australia is that a fail on a temperament test is a guaranteed death sentence. With limited investment in developing skilled foster carer programs, or supported behaviour modification programs, a pound can relieve itself of any obligation to save the pet simply by determining that it ‘failed’ its temperament test. Even if the issue is common and treatable and the prognosis for rehabilitation is good (like cases of ‘jumping’, ‘mouthing’ or ‘resource guarding’) a failed test usually means the pet will lose its life.
Previously there has really only been US research available on the effectiveness of temperament testing in determining a pets future behaviour in the home. Surely, Australia shelter workers are more highly skilled than our US counterparts? Our temperament testings more scientific and advanced? And surely in our pounds, where pets are killed in the thousands, rather than the millions like in the US, pets are given a much more fair evaluation and chance at treatment?
From the Oct-Dec 10 Journal of Applied Animal Welfare Science examined data from 11 shelters and pounds across six states. They also observed 50 shelter-dog assessments and interviewed 26 shelter workers.
Guess how many of the pounds and shelters were using standardized protocols that had been scientifically validated.
Zero.
In fact, only two of the testing protocols had specific and standardized step-by-step instructions on how to administer the assessments. And “there was little consistency in how the protocols were administered, partly because of constraints including limited availability of time and resources”.
When it came to interpreting the results of the test “in most cases there were no guidelines to assist scorers in interpreting the results of the individual assessments. Because the protocols were so diverse internally, it was often impossible to sum scores into a meaningful total. Decisions regarding the fate of assessed dogs were therefore made subjectively.
So if staff have to make a subjective assessment on results, how confident are they that they’ve had enough support and training to accurately asses a dog? Just 35% felt ‘very confident’, while 50% felt ‘somewhat confident’.
“The interviews with shelter staff revealed, notably, that experience was not correlated with confidence in the current assessment protocol or confidence in the respondent’s ability to accurately assess dogs.”
Potentially, therefore, shelter staff are required to make important decisions based on inadequate training, potentially invalid assessment protocols, and subjective interpretations of behaviour in which they have limited confidence; shelter staff members decide which dogs live or die (are euthanized) and which dogs are released into the community. This is unacceptable in terms of placing both dogs and members of the public at risk. It also potentially causes undue stress to shelter staff and may result in high staff turnover and traumatic stress symptoms.
When the outcome of a failed test is death, the lack of true scientific data on temperament tests leaves both staff and animals vulnerable. As we move forward improvements in how we use and deliver behavioural assessments and increasing capacity of treatment programs for pets who enter the pound system will become an integral part of modern sheltering. Killing pets rather than providing options for them to be rehabilitated will no longer be acceptable to pet lovers, shelter donors or the shelter workers themselves.
See also: The Foundation for the Charismatic, Good Looking, Healthy Homeless
