ABSTRACT
Direct observation and measurement of behavior are the cornerstones of effective research and practice in applied behavior analysis (ABA). Despite the importance of data collection, little research is available to guide behavior analysts on issues, such as accuracy, as it relates to data collection integrity. In this paper we examined the extent to which observer reliability is influenced by observer load (i.e., the number of behaviors being simultaneously recorded) and response rate. Results show that both load and response rate may impact the reliability of data collected by newly trained direct care staff. Implications for the design of data collection systems will be discussed, as well as additional considerations and future directions needed on the topic of data collection integrity.