There is a long history of discussion about the importance of data collection methods in the field of applied behavioural analysis (Wolf, 1978). In fact, Baer, Wolf and Risley (1968) described precise measurements as a way to meet the field`s behavioural expectations. In other words, the quantification of behaviour must be done reliably so that changes in behaviour can be documented objectively. Without such reliable quantification, there can be no «credible demonstration of the events that may be responsible for the occurrence or absence of such behaviours» (Baer et al., 1968, p. 94-95). In addition, data are needed because they can affect life-changing interventions (Vollmer, Sloman, and St. Peter Pipkin, 2008). The current study of two experiments evaluated potential interventions that could be presented to participants working in a psychiatric ward with children with intellectual and viable disabilities to improve accuracy in data collection. In Experiment 1, we found that four participants presented four different data models. One participant responded well to the basic data collection measures. The introduction of a clicker resulted in an increase in data collection for another participant.
The other two required a combination of a clicker plus a simplified observation (participant 3) and a combination of these two components and a signal for the end of an interval (participant 4). We then assessed whether four other participants would respond positively to a treatment package consisting of all the intervention elements (experiment 2). We found that there was an increase in IOA for all four participants. Advanced data coders were the first two authors. Each had received behavioural data coding training as part of their course sequence to become certified analysts. In addition, the two data were collected between them and showed a high degree of agreement (i.e. more than 80% approval) in the current study. Data collection is a hallmark of effective behavioural analysis.
Accurate data collection allows a behavioral analyst to assess the effectiveness of behavioural treatment. The current study evaluated the use of a clicker, simplified observation and timer to improve the accuracy of data collection in a psychiatric unit for children with mental and developmental disabilities. Experiment 1, conducted as part of a combined multi-base approach and reverse design, was an evaluation designed to identify a set of interventions for four participants employed in the psychiatric ward. Interventions that resulted in the highest interobservier agreement (IOA) were highly individualized. As a result, we selected the most comprehensive intervention and set up four other participants during Intervention 2. The results showed that this intervention improved the IOA for these additional participants, as evaluated as part of a multi-base participation project.