Reliable measurement requires that behaviors are defined objectively. Vague terms such as anger, depression, aggression or tantrums are redefined in observable and quantifiable terms, so their frequency, duration or other measurable properties can be directly recorded (Sulzer-Azaroff & Mayer, 1991). For example, a goal to reduce a child’s aggressive behavior might define “aggression” as: “attempts, episodes or occurrences (each separated by 10 seconds) of biting, scratching, pinching or pulling hair.” “Initiating social interaction with peers” might be defined as: “looking at classmate and verbalizing an appropriate greeting.”
ABA interventions require a demonstration of the events that are responsible for the occurrence, or non-occurrence, of behavior. ABA uses methods of analysis that yield convincing, reproducible, and conceptually sensible demonstrations of how to accomplish specific behavior changes (Baer & Risley, 1987). Moreover, these behaviors are evaluated within relevant settings such as schools, homes and the community. The use of single case experimental design to evaluate the effectiveness of individualized interventions is an essential component of programs based upon ABA methodologies.