Why do shelters use temperament tests?
Shelter staff use temperament tests to find out if an animal has a behavioral problem which may affect his adoption. After the test is done the staff will be better able to place the animal into an appropriate home, thereby decreasing the amount of animals returned to the shelter after adoption. If the animal is found to have a behavior issue, shelter staff will work with the animal prior to being placed in a home as well as inform the adopter about the behavior. If the behavior is too serious and the prognosis is poor, the shelter may opt to have the animal killed in order to prevent any future bites and liability issues.

Who uses these tests?
Most shelters use some form of temperament testing to evaluate the adoptability of their animals. These tests are used widely by both “kill” and no-kill shelters alike. Leaders in the movement of behavioral assessments include Emily Weiss, PhD, the creator of the “SAFER Test” and Sue Sternberg, the creator of “Assess-a-Pet.”

What kind of tests are used?
There are many temperament tests used today, however, most shelters use either the “SAFER Test,” the “Assess-a-Pet,” or some derivative of these two.

Ideally, each test is performed by shelter staff who have trained specifically to perform the test. The dog will be given a period of adjustment ranging from 3 to 5 days so that he is familiar with shelter life prior to the first test. To begin, the dog will be leashed and scanned for behavioral cues which may be a safety hazard, such as nervousness, shyness, or reactivity. If the dog is too edgy, he may be given a longer period of adjustment.

The actual testing phase has many steps, though the whole process will only take some 15 minutes. The initial steps may test the dog’s reactions to gentle handling. These include repeatedly examining the dogs teeth by lifting up his lips, hugging the dog by a gentle head lock, and gently pinching and pulling at the dog. Further phases test the dog’s reactions to children via a life-size doll and taking the dog’s bowl and treats away with a rubber hand. During such action staff assess play style, level of arousal, and ability to calm down once play has ceased followed by meeting a strange person, dog, or another animal.

Some tests will include a visit with a veterinarian to rule out any medical cause for a particular reaction. The tests may be repeated to re-evaluate the dog after behavior modification efforts or experience in a foster home.

What do the results mean?
Although it varies, most tests are designed to include wide range of possible results. Results may be graded on a scale of A-B-C-D-F, where F is considered failure. The dog is also ranked on a variety of reactions such as: allows hugging; allows but with struggle; allows but becomes tense; intense struggle; tries to bite. Results are then grouped into categories like: no restrictions; home without kids; home without other dogs; unsocial; borderline; failed; no-issue. Dogs who exhibited the most serious issues are typically candidates for euthanasia.

Do these tests work?
That depends on whom you ask. There are very few studies of “behavioral testing” and their results vary considerably. Some researchers have found behavioral differences between breeds, sexes, ages, and whether or not the animal was intact, while others have found just one or none of these differences.

The variety in results may be due to a number of variables. One problem facing researchers is that dogs’ behaviors are largely contextual, yielding different behavioral cues depending on outside influences. Dogs in shelters may have heightened stress levels due to long periods of confinement. Such stress often only worsens in time.

The results may also depend on the perspective of the people involved, be they the evaluators themselves or their adoptive parents. Behaviors considered unacceptable to one person may not be considered unacceptable to the next. The number of dogs who pass or fail these tests can differ depending on whether or not the shelter is “kill” or no-kill and their level of dedication to the cause. Some argue that shelters who are bogged-down with too many dogs and too little space may be inclined to mark more dogs as un-adoptable than those experiencing fewer of these stresses.

Some researchers however, have found that behavioral tests can significantly reduce the number of dogs returned for aggressiveness. A recent research paper, “Behavioral evaluation and demographic information in the assessment of aggressiveness in shelter dogs” by Kelley S. Bollen and Joseph Horowitz, reviewed the effectiveness of the “Assess-a-Pet” test in a shelter environment. During the study, 2017 shelter dogs were tested for aggressive tendencies. The subjects were placed into categories of “failed; unsocial; borderline; no-issue.” After placement in adoptive homes, dogs were evaluated in a 6-month follow-up via a phone interview with the adoptive parent. By implementing the behavioral tests, returns of adoptions fell from 19% the previous year to 14% and returns for aggression fell from 5% to 3.5%. Although these reductions are statistically significant, they are relatively small.

Are these tests fair?
The research mentioned may indicate a reduced return rate, but it did not (and possibly could not) fully evaluate those dogs who failed the test for aggressiveness. 796 dogs failed the evaluation and 759 of these were killed; 112 of those killed showed severe aggression on only one component of the test. It is also worthwhile to note that six dogs were killed who exhibited aggressive behaviors in their previous home, but had passed the evaluation.

“In order to be fair, a temperament test must do two things: (1) screen out aggression and (2) ensure that friendly, scared, shy, sick, or injured dogs do not get wrongly executed. By focusing on the first prong, traditional shelters have ignored the second, a violation that goes to the core of the no-kill ideal: Animals are to be judged and treated as individuals.” Nathan Winograd of The No Kill Advocacy Center.

It is clear that some of the dogs evaluated indeed exhibited aggression in their home, despite having passed the test. While such false-positives do not directly describe the tests’ propensities for false-negatives, we may suppose that dogs who failed the test may not actually have been aggressive. How many of these dogs were wrongfully killed? Do we find it acceptable to wrongfully kill a single dog? Do we find it moral or merciful to kill an animal from decisions solely contingent on a human-perceived behavior? Perhaps these are questions behavioral tests may never fully answer.

Visit The Bark Unleashed for an in-depth study of this topic.