A multiple-baseline across behaviors design was utilized to examine the relation between performance-based feedback (PF) and the frequency of teaching teams' use of target behaviors during observation sessions with six teachers in three inclusive preschool classrooms. Teachers increased their use of target behaviors when feedback was provided and maintained their use over time when feedback was withdrawn; however, there was considerable variability across teachers. Furthermore, teachers did not increase their use of target behaviors during covert observations--when they did not know they were being observed. The current study replicates and extends the research in this area by examining the provision of PF to teaching teams and with complex behaviors (e.g., play expansions, promoting social interactions).