Growing evidence suggests that Teacher-Child Interaction Training-Universal (TCIT-U) is effective for increasing teachers’ use of strategies that promote positive child behavior, but more rigorous research with larger, diverse samples is needed to understand the effects of TCIT-U on teacher and child outcomes in early childhood special education. Using a cluster randomized control trial, we evaluated the effects of TCIT-U on (a) teacher skill acquisition and self-efficacy and (b) child behavior and developmental functioning. Teachers in the TCIT-U group (n = 37) exhibited significantly greater increases in positive attention skills, increased consistent responding, and decreased critical statements relative to teachers in the waitlist control group (n = 36) at post and 1-month follow-up (d’s range from 0.52 to 1.61). Teachers in the TCIT-U group also exhibited significantly fewer directive statements (d’s range from 0.52 to 0.79) and greater increases in self-efficacy compared to waitlist teachers at post (d’s range from 0.60 to 0.76). TCITU was also associated with short-term benefits for child behavior. Frequency (d = 0.41) and total number of behavior problems (d = 0.36) were significantly lower in the TCIT-U group than in the waitlist group at post (but not follow-up), with small-to-medium effects. The waitlist group, but not the TCIT-U group, demonstrated an increasing trend in number of problem behaviors over time. There were no significant between-group differences in developmental functioning. Current findings build support for the effectiveness of TCIT-U as universal prevention of behavior problems with an ethnically and racially diverse sample of teachers and children, including children with developmental disabilities. Implications for implementation of TCIT-U in the early childhood special education setting are discussed. (author abstract)