76 (5) 359-368

Validity of the Socially Acceptable Behavior (SAB) test as a measure of aggression in dogs towards non-familiar humans

For many years dog aggression has been a matter of concern for the authorities and dog breedinggroups. In order to assess aggressive tendencies in dogs, an aggression test was developed (Netto and Planta,1997). This test consisted of 43 subtests, had to be performed in an indoor enclosure and lasted for approximately45 minutes per dog. It seemed impossible to use this test for a large population, therefore theDutch Kennel Club (Raad van Beheer) asked the first author to develop a shorter version of this test thatpreferably should be performed outdoors. The test, called the Socially Acceptable Behavior test (SAB test),consists of only 16 subtests and is performed outdoors. The definition of aggressive biting during the testincludes not only aggressive bites or snaps, but also aggressive attacks in which the dog makes a lunge in anattempt to bite, which fails because it has reached the end of its leash. In this article the validity of the testfor aggressive biting behavior is examined by analyzing the data for 330 dogs. To do so, a comparison wasmade between the test result and the aggressive biting behavior of the dogs prior to the test, as reportedby the owners and confirmed by behavioral consultants. The correspondence between the history of bitingand the aggressive biting behavior during the test is 82%, when no aggressive biting behavior at all is consideredacceptable. When the aggressive biting behavior is allowed in maximum one of 8 specified subtests,in biting behavior is allowed in maximum one of 8 specified, the correspondence raises to 88%, and of the27 false negatives (28.7%) 18 dogs performed only territorial aggression prior to the test.In the second phase, the predictability of aggressive biting behavior towards people, as shown in the test,is calculated by comparing the test results of 220 dogs with the aggressive biting behavior of these dogsshown in a period of at least 1 year after the test, as reported by the owners. The predictability is 81.8%.The test can be considered to predict in a statistically significant way the occurrence of future aggressivebiting behavior of dogs towards unfamiliar people in a non-territorial context in the first year after the test.The existence of false negative results means that the assumption that a dog that passes the test will neverbite later is not correct, as aggression is often very contextual. The value of the individual test is that it supplementsthe other evaluation methods in terms of behavioral consultation and risk evaluation. Furtherinvestigation is necessary to compare a dog’s individual result to its behavior over a longer period later inlife and to investigate the use of the test as a tool in breeding programs.

Full text: 
pp 359-368
In practice