conclusions drawn about program outcomes from testing an intervention as it is actually executed in the context of routine everyday service delivery. This method contains several threats to internal validity because it must rely on procedures that are not scientifically controlled. Compare program efficacy.